rank model variants on average precision, and say when they are not comparable

The workbench ranks model variants by a stored F1, each measured at that
variant's own confidence threshold. A conservatively calibrated detector then
looks worse than a liberal one without detecting anything differently: the
number says as much about the threshold as about the model. POST
/detection/runs/compare ranks on average precision instead, which describes the
whole ranking a model produced, and keeps each run's own-threshold F1 visible
next to it so the difference between the two readings is auditable.

Comparability comes before the ranking. Runs over different source rasters,
scored against different references, without a proven inference footprint, or
covering a different evaluated population are not alternatives to one another,
and no metric makes them so. The report names which of those applies and still
returns the numbers — they are simply not a ranking.

Each run is scored through the same QA path the workbench uses, so a comparison
and the persisted quality checks cannot drift apart.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

backend/app/schemas/detection.py

@@ -78,6 +78,25 @@ class DetectionQaRequest(BaseModel):
     calibration_thresholds: list[float] = Field(default_factory=list, max_length=32)
 
 
+class DetectionComparisonRequest(BaseModel):
+    """Place several runs side by side against one reference."""
+
+    analysis_run_ids: list[UUID] = Field(min_length=2, max_length=12)
+    reference_dataset_id: UUID
+    iou_threshold: float = Field(default=0.5, ge=0.0, le=1.0)
+
+
+class DetectionComparisonResponse(BaseModel):
+    reference_dataset_id: UUID
+    iou_threshold: float
+    # Whether these runs answer the same question at all, and why not if they
+    # do not. Numbers from incomparable runs are reported but never ranked as
+    # if they were alternatives.
+    comparability: dict
+    ranking_metric: str
+    rows: list[dict]
+
+
 class DetectionRunResponse(BaseModel):
     model_config = ConfigDict(protected_namespaces=())