feat: add Dataset Intelligence service to Python SDK

- Add dataset_intelligence_service.py resource with create/append/status/list methods
- Add dataset_intelligence_service.py models for typed request/response
- Wire DI service into resources __init__ and models __init__

Made-with: Cursor

Author: latenceainew

src/latence/_models/__init__.py

@@ -66,6 +66,12 @@
     StageResult,
     StageStatus,
 )
+from .dataset_intelligence_service import (
+    DatasetIntelligenceDeltaSummary,
+    DatasetIntelligenceResponse,
+    DatasetIntelligenceStageTiming,
+    DatasetIntelligenceUsage,
+)
 from .redaction import DetectPIIResponse, RedactionConfig
 
 __all__ = [
@@ -128,6 +134,11 @@
     # Redaction
     "DetectPIIResponse",
     "RedactionConfig",
+    # Dataset Intelligence
+    "DatasetIntelligenceDeltaSummary",
+    "DatasetIntelligenceResponse",
+    "DatasetIntelligenceStageTiming",
+    "DatasetIntelligenceUsage",
     # Jobs
     "CreditsResponse",
     "JobCancelResponse",

src/latence/_models/dataset_intelligence_service.py

@@ -0,0 +1,74 @@
+"""Pydantic models for the Dataset Intelligence service."""
+
+from __future__ import annotations
+
+from typing import Any
+
+from pydantic import BaseModel, ConfigDict, Field
+
+from .common import BaseResponse
+
+
+class DatasetIntelligenceStageTiming(BaseModel):
+    """Timing information for a single pipeline stage."""
+
+    model_config = ConfigDict(extra="allow")
+
+    stage: str = ""
+    elapsed_ms: float = 0.0
+    status: str = "completed"
+
+
+class DatasetIntelligenceUsage(BaseModel):
+    """Credit usage breakdown."""
+
+    model_config = ConfigDict(extra="allow")
+
+    credits: float = 0.0
+    calculation: str = ""
+    details: dict[str, Any] = Field(default_factory=dict)
+
+
+class DatasetIntelligenceDeltaSummary(BaseModel):
+    """Summary of changes in an incremental (append) ingestion."""
+
+    model_config = ConfigDict(extra="allow")
+
+    files_added: int = 0
+    files_updated: int = 0
+    files_unchanged: int = 0
+    new_entities: int = 0
+    merged_entities: int = 0
+    new_edges: int = 0
+    removed_edges: int = 0
+    ontology_types_added: int = 0
+    rotate_retrain: str = ""
+    delta_ratio: float = 0.0
+    train_metrics: dict[str, Any] | None = None
+
+
+class DatasetIntelligenceResponse(BaseResponse):
+    """Response from the Dataset Intelligence service.
+
+    Returned by ``enrich()``, ``build_graph()``, ``build_ontology()``,
+    and ``run()`` on the SDK resource.
+    """
+
+    model_config = ConfigDict(extra="allow")
+
+    endpoint_id: str = "dataset_intelligence"
+    tier: str = ""
+    dataset_id: str = ""
+    mode: str = "create"
+    data: dict[str, Any] = Field(default_factory=dict)
+    usage: DatasetIntelligenceUsage = Field(
+        default_factory=DatasetIntelligenceUsage
+    )
+    stage_timings: list[DatasetIntelligenceStageTiming] = Field(
+        default_factory=list
+    )
+    delta_summary: DatasetIntelligenceDeltaSummary | None = None
+    processing_time_ms: float = 0.0
+    version: str = ""
+    error: str | None = None
+    error_code: str | None = None

src/latence/resources/__init__.py

@@ -2,14 +2,18 @@
 
 from __future__ import annotations
 
+from .chunking import AsyncChunking, Chunking
 from .colbert import AsyncColBERT, ColBERT
 from .colpali import AsyncColPali, ColPali
 from .compression import AsyncCompression, Compression
 from .credits import AsyncCredits, Credits
+from .dataset_intelligence_service import (
+    AsyncDatasetIntelligenceService,
+    DatasetIntelligenceService,
+)
 from .document_intelligence import AsyncDocumentIntelligence, DocumentIntelligence
 from .embed import AsyncEmbed, Embed
 from .embedding import AsyncEmbedding, Embedding
-from .chunking import AsyncChunking, Chunking
 from .enrichment import AsyncEnrichment, Enrichment
 from .experimental import AsyncExperimentalNamespace, ExperimentalNamespace
 from .extraction import AsyncExtraction, Extraction
@@ -25,6 +29,7 @@
     "ColPali",
     "Compression",
     "Credits",
+    "DatasetIntelligenceService",
     "DocumentIntelligence",
     "Embed",
     "Embedding",
@@ -41,6 +46,7 @@
     "AsyncColPali",
     "AsyncCompression",
     "AsyncCredits",
+    "AsyncDatasetIntelligenceService",
     "AsyncDocumentIntelligence",
     "AsyncEmbed",
     "AsyncEmbedding",