feat: Add hybrid quality evaluation system with CLIP and VLM support

- Add FeatureExtractor for CLIP-based image/text feature extraction
- Add ObjectiveMetricsCalculator for technical quality metrics
- Add VLMEvaluator for vision language model evaluation
- Add HybridQualityGate combining objective + VLM evaluation
- Enhance CharacterMemory with visual feature support
- Add quality optional dependency (torch, ftfy, regex)
- Add unit tests for new modules

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

tests/test_objective_metrics.py

@@ -0,0 +1,64 @@
+# Copyright (C) 2025 AIDC-AI
+# Tests for ObjectiveMetricsCalculator
+
+import pytest
+from pathlib import Path
+
+from pixelle_video.services.quality.objective_metrics import (
+    ObjectiveMetricsCalculator,
+    TechnicalMetrics,
+)
+
+
+class TestTechnicalMetrics:
+    """Tests for TechnicalMetrics dataclass"""
+
+    def test_default_values(self):
+        metrics = TechnicalMetrics()
+        assert metrics.sharpness_score == 0.0
+        assert metrics.overall_technical == 0.0
+        assert metrics.issues == []
+
+    def test_to_dict(self):
+        metrics = TechnicalMetrics(
+            sharpness_score=0.8,
+            brightness_score=0.5,
+            issues=["test issue"]
+        )
+        d = metrics.to_dict()
+        assert d["sharpness_score"] == 0.8
+        assert "test issue" in d["issues"]
+
+
+class TestObjectiveMetricsCalculator:
+    """Tests for ObjectiveMetricsCalculator"""
+
+    def test_init_default(self):
+        calc = ObjectiveMetricsCalculator()
+        assert calc.sharpness_threshold == 0.3
+
+    def test_init_custom(self):
+        calc = ObjectiveMetricsCalculator(sharpness_threshold=0.5)
+        assert calc.sharpness_threshold == 0.5
+
+    def test_analyze_nonexistent_image(self):
+        calc = ObjectiveMetricsCalculator()
+        metrics = calc.analyze_image("/nonexistent/path.png")
+        assert len(metrics.issues) > 0
+        assert "failed" in metrics.issues[0].lower()
+
+    def test_analyze_real_image(self, tmp_path):
+        """Test with a real image file"""
+        from PIL import Image
+
+        # Create test image
+        img = Image.new("RGB", (256, 256), color=(128, 128, 128))
+        img_path = tmp_path / "test.png"
+        img.save(img_path)
+
+        calc = ObjectiveMetricsCalculator()
+        metrics = calc.analyze_image(str(img_path))
+
+        assert 0.0 <= metrics.sharpness_score <= 1.0
+        assert 0.0 <= metrics.brightness_score <= 1.0
+        assert 0.0 <= metrics.overall_technical <= 1.0