make QA scoring reproducible and threshold-independent

The greedy IoU matcher gave a reference to whichever candidate was offered
first. Row order decided that, and every detection in a run shares one
transaction timestamp, so ordering by created_at left the assignment
undefined: the same QA run over the same data produced different mean IoU,
and the geometry shown to a reviewer as a false positive could be the better
of two detections. Candidates are now ranked by confidence with feature
identity as tiebreaker, which is also the COCO/PASCAL rule.

A single precision/recall/F1 triple describes one operating point, so two
models cannot be compared from it: a conservatively calibrated model looks
worse at a low confidence cut and better at a high one without detecting
anything differently. DetectionMetricsService adds the full curve, average
precision and the threshold where F1 actually peaks.

Also:
- report the population the metrics were computed over, so
  matches + false_positives equals candidate_feature_count even under an
  area filter; raw dataset totals move to the _raw fields;
- state whether candidates are axis-aligned boxes or footprint polygons.
  A box can never reach IoU 1 against a rotated building, so the strict
  score has a ceiling that has nothing to do with detection quality.

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

backend/app/schemas/qa.py

@@ -19,8 +19,14 @@ class QaProviderComparisonRequest(BaseModel):
 class QaProviderComparisonResult(BaseModel):
     status: str
     warnings: list[str] = Field(default_factory=list)
+    # Counts of the population that was actually matched, so that
+    # ``matches + false_positives == candidate_feature_count`` holds even when
+    # an area filter or an unparseable geometry removed features. The ``_raw``
+    # fields keep the untouched dataset totals visible next to them.
     candidate_feature_count: int
     reference_feature_count: int
+    candidate_feature_count_raw: int | None = None
+    reference_feature_count_raw: int | None = None
     matches: int
     false_positives: int
     false_negatives: int