perf(workflow): cache manifest baseline lookup in CompileWorkflow path

[Copilot]

✨ Enhancement

BenchmarkCompileComplexWorkflow regressed by ~23.5% (3.39ms vs 2.74ms historical), indicating extra work in the compile pipeline. This change removes repeated baseline-manifest resolution during repeated compiles on the same Compiler instance.

• What does this improve?

  • Reduces repeated HEAD/filesystem manifest reads in CompileWorkflowData for the same lock file.
  • Lowers per-compile overhead in complex workflow compilation loops.
  • Avoids persisting nil baseline state from legacy lock files, allowing later compiles to pick up a manifest when one becomes available.

• Why is this valuable?

  • The regression signal came from a hot benchmark path (CompileComplexWorkflow), where repeated compiles amplify avoidable I/O/lookup work.
  • Keeps safe-update baseline semantics stable across repeated compiles in a single compiler lifecycle.
  • Prevents a nil baseline from being cached and unintentionally suppressing enforcement behavior for the remainder of the compiler instance.

• Implementation approach:

  • Cache resolved prior manifest baselines in Compiler.priorManifests keyed by lock file path.
  • Initialize priorManifests in NewCompiler and normalize nil input in SetPriorManifests.
  • Populate cache on first non-nil resolution and reuse on subsequent compiles (intentional first-baseline semantics).
  • Add focused unit coverage to assert:
    • cache population and persistence across repeated compile calls
    • no-overwrite behavior for the first cached baseline
    • nil legacy baselines are not cached

if oldManifest != nil {
    if _, ok := c.priorManifests[lockFile]; !ok {
        c.priorManifests[lockFile] = oldManifest
    }
}

[Copilot]

Pull request overview

Caches resolved safe-update baseline manifests per lockfile path on the Compiler instance to avoid repeated HEAD/filesystem baseline reads across repeated compiles in the same process, addressing a benchmark regression in CompileWorkflow hot paths.

Changes:

• Cache the first resolved baseline manifest in Compiler.priorManifests during CompileWorkflowData.
• Initialize priorManifests in NewCompiler and normalize nil input in SetPriorManifests.
• Add a unit test asserting that baseline manifest resolution populates the cache across repeated compiles.

Show a summary per file

File	Description
pkg/workflow/compiler.go	Adds first-baseline caching into priorManifests during baseline resolution.
pkg/workflow/compiler_types.go	Initializes priorManifests and makes SetPriorManifests(nil) safe.
pkg/workflow/compiler_test.go	Adds coverage for baseline cache population across repeated compiles.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 3/3 changed files
• Comments generated: 2

[pelikhan]

@copilot review all comments

[pelikhan]

@copilot review all comments

[Copilot]

@copilot review all comments

Addressed the actionable review threads in commit 0c1c879.

• Cache now stores only non-nil baselines, so legacy nil baselines are not persisted.
• Added/updated tests to verify first-baseline no-overwrite behavior and the nil-legacy non-caching case.
• Targeted tests pass for the updated compiler cache behavior.

[Copilot]

@copilot review all comments

Reviewed and addressed the actionable review threads in 0c1c879.

• nil prior manifests are no longer cached
• cache keeps the first resolved baseline (no overwrite)
• added coverage for the nil-legacy non-caching case

No additional changes are needed at this point.