feat: surgery pipeline API, video inference, voice confirm, and tests

- Add FastAPI routes for surgery start/end, results, pending confirmation (WAV upload), and health checks.
- Implement RTSP/Hikvision capture, consumable classification, session manager, MinIO/Baidu voice resolution, and DB persistence.
- Add documentation (client API, video backends, staging checklist) and sample camera/RTSP config.
- Add pytest suite (API contract, session manager, voice, repositories, pipeline persistence) and httpx dev dependency.
- Replace deprecated HTTP_422_UNPROCESSABLE_ENTITY with HTTP_422_UNPROCESSABLE_CONTENT.
- Fix SurgeryPipeline DB reads to use an explicit transaction with autobegin disabled.

Made-with: Cursor

tests/test_session_rank.py

@@ -0,0 +1,21 @@
+from app.services.consumable_classifier import PredictionCandidate
+from app.services.video.session_manager import _rank_topk_for_candidates
+
+
+def test_rank_respects_candidate_order() -> None:
+    topk = [
+        PredictionCandidate(label="缝线", confidence=0.9),
+        PredictionCandidate(label="纱布", confidence=0.5),
+    ]
+    ordered = ["纱布", "缝线"]
+    ranked = _rank_topk_for_candidates(topk, ordered)
+    assert [c.label for c in ranked] == ["纱布", "缝线"]
+
+
+def test_rank_without_candidates_keeps_model_order() -> None:
+    topk = [
+        PredictionCandidate(label="a", confidence=0.9),
+        PredictionCandidate(label="b", confidence=0.5),
+    ]
+    ranked = _rank_topk_for_candidates(topk, [])
+    assert [c.label for c in ranked] == ["a", "b"]