feat: Phase 2.6 edges migration to Qdrant (TKG-only architecture)

Phase 2.6.1: co_occurrence_edges migration
- build_co_occurrence_edges_from_qdrant()
- Qdrant embeddings → frame grouping → YOLO objects
- Result: 6679 edges (vs 6701 PostgreSQL)

Phase 2.6.2: face_face_edges migration
- build_face_face_edges_from_qdrant()
- Qdrant embeddings → frame grouping → face pairs
- mutual_gaze detection preserved
- Result: 6 edges (exact match)

Phase 2.6.3: speaker_face_edges migration
- build_speaker_face_edges_from_qdrant()
- Qdrant embeddings → trace_id frame ranges
- SPEAKS_AS edge creation

Architecture:
- All edges use Qdrant payload (no face_detections queries)
- PostgreSQL fallback for empty Qdrant
- Estimated 3.6x performance improvement

Testing:
- Playground (3003): ✓ All Phase 2.6 logs verified
- Edge counts: ✓ Close match with PostgreSQL
- Fallback: ✓ Working

Docs:
- docs_v1.0/DESIGN/TKG_PHASE2_6_EDGES_MIGRATION.md
- docs_v1.0/M4_workspace/2026-06-21_phase2_6_test.md

v1.1/requirements_v1.11.txt

@@ -0,0 +1,33 @@
+# Momentry Core - Python Dependencies
+# Locked versions - Generated from venv pip freeze
+# Install: /opt/homebrew/bin/python3.11 -m pip install -r requirements.txt
+
+# Core - Video Processing
+opencv-python==4.13.0.92
+
+# Redis - Message Bus
+redis==7.3.0
+
+# ASR - Speech Recognition
+torch==2.8.0
+torchvision==0.23.0
+torchaudio==2.8.0
+faster-whisper==1.2.1
+whisperx==3.8.2
+
+# Scene Detection
+scenedetect==0.6.7.1
+
+# Object Detection & Pose
+ultralytics==8.4.23
+
+# OCR - Text Recognition
+easyocr==1.7.2
+
+# Additional dependencies (automatically installed)
+# pillow>=12.0.0
+# numpy>=2.0.0
+# pandas>=3.0.0
+
+# Dev - Code Quality
+ruff==0.11.2