feat: add dual-track API endpoints for document processing

- Add ProcessingTrackEnum, ProcessingOptions, ProcessingMetadata schemas
- Add DocumentAnalysisResponse for document type detection
- Update /start endpoint with dual-track query parameters
- Add /analyze endpoint for document type detection with confidence scores
- Add /metadata endpoint for processing track information
- Add /download/unified endpoint for UnifiedDocument format export
- Update tasks.md to mark Section 6 API updates as completed

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

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

backend/app/schemas/task.py

@@ -16,6 +16,14 @@ class TaskStatusEnum(str, Enum):
     FAILED = "failed"
 
 
+class ProcessingTrackEnum(str, Enum):
+    """Processing track enumeration for dual-track processing"""
+    OCR = "ocr"           # PaddleOCR PP-StructureV3 for scanned documents
+    DIRECT = "direct"     # PyMuPDF for editable PDFs
+    HYBRID = "hybrid"     # Mixed processing
+    AUTO = "auto"         # Auto-detect best track
+
+
 class TaskCreate(BaseModel):
     """Task creation request"""
     filename: Optional[str] = Field(None, description="Original filename")
@@ -117,3 +125,64 @@ class UploadResponse(BaseModel):
     file_size: int = Field(..., description="File size in bytes")
     file_type: str = Field(..., description="File MIME type")
     status: TaskStatusEnum = Field(..., description="Initial task status")
+
+
+# ===== Dual-Track Processing Schemas =====
+
+class ProcessingOptions(BaseModel):
+    """Processing options for dual-track OCR"""
+    use_dual_track: bool = Field(default=True, description="Enable dual-track processing")
+    force_track: Optional[ProcessingTrackEnum] = Field(None, description="Force specific track (ocr/direct)")
+    language: str = Field(default="ch", description="OCR language code")
+    include_layout: bool = Field(default=True, description="Include layout analysis")
+    include_images: bool = Field(default=True, description="Extract and save images")
+    confidence_threshold: Optional[float] = Field(None, ge=0, le=1, description="OCR confidence threshold")
+
+
+class AnalyzeRequest(BaseModel):
+    """Document analysis request"""
+    use_dual_track: bool = Field(default=True, description="Enable dual-track processing")
+    force_track: Optional[ProcessingTrackEnum] = Field(None, description="Force specific track")
+    language: str = Field(default="ch", description="OCR language")
+    include_layout: bool = Field(default=True, description="Include layout analysis")
+
+
+class DocumentAnalysisResponse(BaseModel):
+    """Document type analysis response"""
+    task_id: str
+    filename: str
+    recommended_track: ProcessingTrackEnum
+    confidence: float = Field(..., ge=0, le=1, description="Detection confidence")
+    reason: str = Field(..., description="Reason for recommendation")
+    document_info: dict = Field(default_factory=dict, description="Document metadata")
+    is_editable: bool = Field(..., description="Whether document has extractable text")
+    text_coverage: Optional[float] = Field(None, description="Percentage of text coverage")
+    page_count: Optional[int] = Field(None, description="Number of pages")
+
+
+class ProcessingMetadata(BaseModel):
+    """Processing metadata included in responses"""
+    processing_track: ProcessingTrackEnum
+    processing_time_seconds: float
+    language: str
+    page_count: int
+    total_elements: int
+    total_text_regions: int
+    total_tables: int
+    total_images: int
+    average_confidence: Optional[float] = None
+    unified_format: bool = True
+
+
+class TaskResponseWithMetadata(TaskResponse):
+    """Extended task response with processing metadata"""
+    processing_track: Optional[ProcessingTrackEnum] = None
+    processing_metadata: Optional[ProcessingMetadata] = None
+
+
+class ExportOptions(BaseModel):
+    """Export format options"""
+    format: str = Field(default="json", description="Export format: json, markdown, pdf, unified")
+    include_metadata: bool = Field(default=True, description="Include processing metadata")
+    include_statistics: bool = Field(default=True, description="Include document statistics")
+    legacy_format: bool = Field(default=False, description="Use legacy JSON format for compatibility")