feat(ai-services): add AVM v2 A/B comparison endpoint and tests

Add POST /avm/v2/compare-v1 endpoint that runs both v1 (single-model)
and v2 (ensemble) AVM predictions on the same property and returns a
side-by-side comparison with price diff, confidence delta, and a
recommendation on which model to prefer.

- ABComparisonRequest/Response schemas in avm_v2 models
- compare_v1() method in AVMv2EnsembleService
- 4 new integration tests for the comparison endpoint
- All 47 Python tests pass

Co-Authored-By: Paperclip <noreply@paperclip.ing>

libs/ai-services/app/models/avm_v2.py

@@ -183,3 +183,64 @@ class AVMv2ModelInfo(BaseModel):
     metrics: dict
     is_active: bool = Field(True)
     ab_test_traffic_pct: float = Field(0.0, ge=0, le=1)
+
+
+class AVMv1Summary(BaseModel):
+    """Compact summary of a v1 prediction for comparison."""
+
+    estimated_price_vnd: float
+    confidence: float
+    price_per_m2: float
+    price_range_low: float
+    price_range_high: float
+
+
+class AVMv2Summary(BaseModel):
+    """Compact summary of a v2 prediction for comparison."""
+
+    estimated_price_vnd: float
+    confidence: float
+    price_per_m2_vnd: float
+    price_range_low_vnd: float
+    price_range_high_vnd: float
+    model_version: str
+    ensemble_method: str
+
+
+class ABComparisonRequest(BaseModel):
+    """Request for A/B comparison between v1 and v2."""
+
+    district: str = Field(..., min_length=1)
+    city: str = Field(..., min_length=1)
+    property_type: str = Field(...)
+    area_m2: float = Field(..., gt=0)
+    rooms: int = Field(0, ge=0)
+    bedrooms: int = Field(0, ge=0, description="Alias for rooms, used by v1")
+    floors: int = Field(0, ge=0)
+    frontage: float = Field(0.0, ge=0)
+    has_legal_paper: bool = Field(True)
+    # v2-specific features (optional, defaults applied)
+    distance_to_cbd_km: float = Field(0.0, ge=0)
+    distance_to_metro_km: float = Field(0.0, ge=0)
+    flood_zone_risk: float = Field(0.0, ge=0, le=1)
+    building_age_years: int = Field(0, ge=0)
+    has_elevator: bool = Field(False)
+    has_parking: bool = Field(False)
+    has_pool: bool = Field(False)
+    renovation_score: float = Field(0.5, ge=0, le=1)
+    view_quality: float = Field(0.5, ge=0, le=1)
+    interior_quality: float = Field(0.5, ge=0, le=1)
+    month: int = Field(1, ge=1, le=12)
+    quarter: int = Field(1, ge=1, le=4)
+    is_year_end: bool = Field(False)
+
+
+class ABComparisonResponse(BaseModel):
+    """Side-by-side A/B comparison of v1 vs v2 predictions."""
+
+    v1: AVMv1Summary
+    v2: AVMv2Summary
+    price_diff_vnd: float = Field(..., description="v2 - v1 price difference")
+    price_diff_pct: float = Field(..., description="Percentage difference ((v2-v1)/v1 * 100)")
+    confidence_diff: float = Field(..., description="v2 - v1 confidence difference")
+    recommendation: str = Field(..., description="Which model to prefer and why")