refactor(webapp): drop what the profile never rendered

`PracticeTrendPanel` had no caller outside its own story and test: the detail
page shows `PracticeTrendChip` and nothing else. Its 214 lines went, and with
them `formatTrendGap` and `formatTrendCoverage`, which nothing else called, and
the `barClassName` and `trendPolarity` fields its stacked bar was the only
reader of. `isVerdictStanding` and `ringClass` were exported and never read.

`ReviewRunTimeline` also carried an empty state its only caller made
unreachable — and a worse one: it said "matches these filters" whether or not a
filter was set, while the caller already tells the two apart. The rule moves
into a comment, and the case the deleted test named is now asserted where the
decision is actually made.

The trend provenance sentence counts "pieces of reviewed work" rather than
"work items", which is what `practice-feedback-language.md` calls the unit.

Also fixes the operator note on the practice-standing changeset, which told
callers to replace `/practice-groups` with `/practice-groups`. The retired path
is `/practice-areas`.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MAQxdMw579siMh1wXKEEh4

Author: Valentin Grüner

.changeset/practice-standing-feedback-api.md

@@ -4,4 +4,4 @@
 
 The application API can now answer where a developer stands in each practice group: the current standing and its guidance, how the group has developed across recently reviewed work, which kinds of work contributed feedback, a filterable observation history, and the complete review runs behind it. An undecided observation remains visible in that history without being presented as a verdict. Developers can also replace or delete their response to delivered feedback, recording whether it was helpful, how they handled it, and an optional explanation. Every endpoint answers only for the signed-in developer.
 
-**Operators:** direct API callers must replace `/practice-groups` with `/practice-groups`, use the corresponding group schema and field names, replace `/practices/developer` with `/practices/reviewed`, and move from the older reaction endpoint to the combined response endpoint. Existing response history is preserved; `MIGRATION.md` lists the contract changes. The generated Hephaestus web client is updated in the same release.
+**Operators:** direct API callers must replace `/practice-areas` with `/practice-groups`, use the corresponding group schema and field names, replace `/practices/developer` with `/practices/reviewed`, and move from the older reaction endpoint to the combined response endpoint. Existing response history is preserved; `MIGRATION.md` lists the contract changes. The generated Hephaestus web client is updated in the same release.

webapp/src/components/profile/PracticeGroupDetailPage.test.tsx

@@ -57,6 +57,27 @@ describe("PracticeGroupDetailPage", () => {
 		expect(onSelectPractice).toHaveBeenCalledWith("small-changes");
 	});
 
+	it("tells an empty filtered feed apart from a group that was never reviewed", () => {
+		// One empty feed, two meanings. Saying "no review runs match your filters" to someone who set
+		// none reads as a fault in the page; saying "none yet" to someone filtering hides their filter.
+		const { rerender } = render(
+			<PracticeGroupDetailPage group={group} reviewRuns={[]} isLoading={false} />,
+		);
+		screen.getByText("Review runs appear here once your work has been reviewed.");
+
+		rerender(
+			<PracticeGroupDetailPage
+				group={group}
+				reviewRuns={[]}
+				reviewRunFilters={{ sources: ["scm.pull_request"], severities: [] }}
+				isLoading={false}
+			/>,
+		);
+		screen.getByText(
+			"No review runs match the current filters. Clear them to see every review in this group.",
+		);
+	});
+
 	it("offers navigation when the group is missing", () => {
 		const onBack = vi.fn();
 		render(<PracticeGroupDetailPage isLoading={false} onBack={onBack} />);

webapp/src/components/profile/ReviewObservationRow.tsx

@@ -15,8 +15,8 @@ import { Button } from "@/components/ui/button";
 import { Collapsible, CollapsibleContent, CollapsibleTrigger } from "@/components/ui/collapsible";
 import { Skeleton } from "@/components/ui/skeleton";
 import { cn } from "@/lib/utils";
-import { EvidenceFileBlock } from "./EvidenceFileBlock";
 import { toEvidenceLocations } from "./evidence";
+import { EvidenceFileBlock } from "./EvidenceFileBlock";
 import {
 	OBSERVATION_OUTCOME_PRESENTATION,
 	type ObservationOutcome,

webapp/src/components/profile/ReviewRunTimeline.tsx

@@ -1,6 +1,6 @@
 import type { PracticeGroupReviewObservation, PracticeGroupReviewRun } from "@/api/types.gen";
-import { ReviewRunCard } from "./ReviewRunCard";
 import type { FeedbackUsefulness, ObservationDetailState } from "./review-runs";
+import { ReviewRunCard } from "./ReviewRunCard";
 
 export type { FeedbackUsefulness, ObservationDetailState } from "./review-runs";
 
@@ -24,14 +24,8 @@ export function ReviewRunTimeline({
 	onChangeUsefulness,
 	pendingFeedbackId,
 }: ReviewRunTimelineProps) {
-	if (runs.length === 0) {
-		return (
-			<div className="rounded-lg border border-dashed p-6 text-center text-sm text-muted-foreground">
-				No reviewed work matches these filters.
-			</div>
-		);
-	}
-
+	// No empty state here: the caller decides what an empty feed means — no runs at all, or none
+	// matching the filters — and renders that itself. A second one here could only ever be wrong.
 	return (
 		<ol className="flex min-w-0 flex-col" aria-label="Review runs">
 			{runs.map((run) => (

webapp/src/components/profile/observation-outcome.test.ts

@@ -15,12 +15,14 @@ describe("observation outcome contract", () => {
 		// the practice did not apply to this work, INCONCLUSIVE means it did and the reviewer could not
 		// tell. Collapsing them would report "not assessed" for a practice that WAS looked at.
 		expect(observationOutcome({ presence: "INCONCLUSIVE" })).toBe("INCONCLUSIVE");
-		expect(OBSERVATION_OUTCOME_PRESENTATION.INCONCLUSIVE.trendPolarity).toBeNull();
+		expect(OBSERVATION_OUTCOME_PRESENTATION.INCONCLUSIVE.label).not.toBe(
+			OBSERVATION_OUTCOME_PRESENTATION.NOT_APPLICABLE.label,
+		);
 	});
 
 	it("presents every outcome, including the one the outcome vector does not count", () => {
 		// The first five mirror the server's OutcomeVector 1:1. INCONCLUSIVE is display-only — it is shown
-		// to the developer but contributes no cell to the trend, which is why it carries no polarity.
+		// to the developer but contributes no cell to the trend.
 		expect(Object.keys(OBSERVATION_OUTCOME_PRESENTATION)).toStrictEqual([
 			"PRESENT_GOOD",
 			"ABSENT_GOOD",

webapp/src/components/profile/observation-outcome.ts

@@ -18,46 +18,31 @@ export const OBSERVATION_OUTCOME_PRESENTATION = {
 	PRESENT_GOOD: {
 		label: "Strength shown",
 		className: "text-success",
-		barClassName: "bg-success",
-		trendPolarity: 1,
 	},
 	ABSENT_GOOD: {
 		label: "Risk avoided",
 		className: "text-success",
-		barClassName: "bg-success/45",
-		trendPolarity: 1,
 	},
 	PRESENT_BAD: {
 		label: "Problem observed",
 		className: "text-destructive",
-		barClassName: "bg-destructive",
-		trendPolarity: -1,
 	},
 	ABSENT_BAD: {
 		label: "Expected practice missing",
 		className: "text-destructive",
-		barClassName: "bg-destructive/45",
-		trendPolarity: -1,
 	},
 	NOT_APPLICABLE: {
 		label: "Not assessed",
 		className: "text-muted-foreground",
-		barClassName: "bg-muted-foreground/30",
-		trendPolarity: null,
 	},
 	// A distinct silence from NOT_APPLICABLE: the reviewer looked at work that DID offer the opportunity
 	// and could not claim either way. Collapsing the two would report "no opportunity" for a practice the
 	// reviewer was simply unsure about.
 	INCONCLUSIVE: {
 		label: "Not certain enough to say",
 		className: "text-muted-foreground",
-		barClassName: "bg-muted-foreground/30",
-		trendPolarity: null,
 	},
-} as const satisfies Record<
-	ObservationOutcome,
-	{ label: string; className: string; barClassName: string; trendPolarity: 1 | -1 | null }
->;
+} as const satisfies Record<ObservationOutcome, { label: string; className: string }>;
 
 /**
  * Derives the complete 2×2 observation state plus the two verdict-less states. The API guarantees no

webapp/src/components/profile/practice-trend-presentation.test.ts

@@ -1,10 +1,6 @@
 import { describe, expect, it } from "vitest";
 import type { TrendSupport } from "@/api/types.gen";
-import {
-	formatTrendCoverage,
-	formatTrendGap,
-	formatTrendProvenance,
-} from "./practice-trend-presentation";
+import { formatTrendProvenance } from "./practice-trend-presentation";
 
 const support = (overrides: Partial<TrendSupport> = {}): TrendSupport => ({
 	currentOpportunities: 4,
@@ -19,7 +15,7 @@ const support = (overrides: Partial<TrendSupport> = {}): TrendSupport => ({
 describe("practice trend copy", () => {
 	it("formats a complete comparison from support", () => {
 		expect(formatTrendProvenance(support({ calendarSpanDays: 9 }))).toBe(
-			"Compared your latest 4 reviewed work items with the 4 before them, spanning 9 days.",
+			"Compared your latest 4 pieces of reviewed work with the 4 before them, spanning 9 days.",
 		);
 	});
 
@@ -28,31 +24,11 @@ describe("practice trend copy", () => {
 			formatTrendProvenance(
 				support({ currentOpportunities: 1, previousOpportunities: 0, calendarSpanDays: 1 }),
 			),
-		).toBe("Based on 1 reviewed work item, spanning 1 day.");
+		).toBe("Based on 1 piece of reviewed work, spanning 1 day.");
 		expect(
 			formatTrendProvenance(
 				support({ currentOpportunities: 0, previousOpportunities: 0, calendarSpanDays: undefined }),
 			),
-		).toBe("No reviewed work items are available yet.");
-	});
-
-	it("formats singular, plural, and zero comparison gaps", () => {
-		expect(formatTrendGap(support({ opportunitiesUntilComparable: 1 }))).toBe(
-			"1 more reviewed work item will make a comparison possible.",
-		);
-		expect(formatTrendGap(support({ opportunitiesUntilComparable: 2 }))).toBe(
-			"2 more reviewed work items will make a comparison possible.",
-		);
-		expect(formatTrendGap(support())).toBe("A comparison is available.");
-	});
-
-	it("formats group coverage and omits it for a practice", () => {
-		expect(formatTrendCoverage(support({ comparablePractices: 3, eligiblePractices: 5 }))).toBe(
-			"3 of 5 practices in this group had comparable evidence.",
-		);
-		expect(formatTrendCoverage(support({ comparablePractices: 1, eligiblePractices: 1 }))).toBe(
-			"1 of 1 practice in this group had comparable evidence.",
-		);
-		expect(formatTrendCoverage(support())).toBeUndefined();
+		).toBe("No reviewed work is available yet.");
 	});
 });

webapp/src/components/profile/practice-trend-presentation.ts

@@ -40,39 +40,24 @@ export const PRACTICE_TREND_PRESENTATION = {
 	},
 } as const satisfies Record<TrendDirection, TrendPresentation>;
 
-function reviewedItems(count: number): string {
-	return `${count} reviewed ${count === 1 ? "work item" : "work items"}`;
+/** The counted unit is a *piece of reviewed work* — `practice-feedback-language.md` names it. */
+function reviewedWork(count: number): string {
+	return `${count} ${count === 1 ? "piece" : "pieces"} of reviewed work`;
 }
 
 /** Formats visible provenance from the returned evidence support, never from an assumed time bin. */
 export function formatTrendProvenance(support: TrendSupport): string {
 	const current = support.currentOpportunities;
 	const previous = support.previousOpportunities;
 	const total = current + previous;
-	if (total === 0) return "No reviewed work items are available yet.";
+	if (total === 0) return "No reviewed work is available yet.";
 
 	const span = support.calendarSpanDays;
 	const spanText = span ? `, spanning ${span} ${span === 1 ? "day" : "days"}` : "";
 	if (previous === 0) {
-		return `Based on ${reviewedItems(current)}${spanText}.`;
+		return `Based on ${reviewedWork(current)}${spanText}.`;
 	}
-	return `Compared your latest ${reviewedItems(current)} with the ${previous} before ${
+	return `Compared your latest ${reviewedWork(current)} with the ${previous} before ${
 		previous === 1 ? "it" : "them"
 	}${spanText}.`;
 }
-
-/** Formats the unmet evidence precondition as a concrete next milestone. */
-export function formatTrendGap(support: TrendSupport): string {
-	const missing = support.opportunitiesUntilComparable;
-	if (missing === 0) return "A comparison is available.";
-	return `${missing} more reviewed ${missing === 1 ? "work item" : "work items"} will make a comparison possible.`;
-}
-
-/** Formats group coverage; practice-scoped support has no coverage sentence. */
-export function formatTrendCoverage(support: TrendSupport): string | undefined {
-	const comparable = support.comparablePractices;
-	const eligible = support.eligiblePractices;
-	if (comparable === undefined || eligible === undefined) return undefined;
-	if (eligible === 0) return "No practices are eligible for comparison in this group.";
-	return `${comparable} of ${eligible} ${eligible === 1 ? "practice" : "practices"} in this group had comparable evidence.`;
-}

webapp/src/routes/_authenticated/w/$workspaceSlug/user/$username/practice-groups.$groupSlug.tsx

@@ -20,6 +20,7 @@ import {
 } from "@/components/profile/PracticeGroupDetailPage";
 import type { ObservationDetailState } from "@/components/profile/review-runs";
 import { useAuth } from "@/integrations/auth/AuthContext";
+import { loadedPages } from "@/integrations/tanstack-query/spring-page";
 import { problemDetailOf } from "@/lib/problem-detail";
 
 const ACTIVITY_PAGE_SIZE = 10;
@@ -144,7 +145,7 @@ function PracticeGroupDetail() {
 			return [practice.slug, nextStep];
 		}),
 	);
-	const reviewRuns = activityQuery.data.pages.flatMap((page) => page.content);
+	const reviewRuns = loadedPages(activityQuery.data).flatMap((page) => page.content);
 
 	const observationDetail: ObservationDetailState | undefined = openObservationId
 		? {