feat: Add AI report generation with DIFY integration

- Add Users table for display name resolution from AD authentication
- Integrate DIFY AI service for report content generation
- Create docx assembly service with image embedding from MinIO
- Add REST API endpoints for report generation and download
- Add WebSocket notifications for generation progress
- Add frontend UI with progress modal and download functionality
- Add integration tests for report generation flow

Report sections (Traditional Chinese):
- 事件摘要 (Summary)
- 時間軸 (Timeline)
- 參與人員 (Participants)
- 處理過程 (Resolution Process)
- 目前狀態 (Current Status)
- 最終處置結果 (Final Resolution)

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

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

app/modules/report_generation/schemas.py

@@ -0,0 +1,105 @@
+"""Pydantic schemas for report generation API
+
+Request and response models for the report generation endpoints.
+"""
+from pydantic import BaseModel, Field
+from typing import Optional, List
+from datetime import datetime
+from enum import Enum
+
+
+class ReportStatus(str, Enum):
+    """Report generation status"""
+    PENDING = "pending"
+    COLLECTING_DATA = "collecting_data"
+    GENERATING_CONTENT = "generating_content"
+    ASSEMBLING_DOCUMENT = "assembling_document"
+    COMPLETED = "completed"
+    FAILED = "failed"
+
+
+# Request Schemas
+class ReportGenerateRequest(BaseModel):
+    """Request to generate a report (optional parameters)"""
+    include_images: bool = Field(default=True, description="Whether to embed images in the report")
+    include_file_list: bool = Field(default=True, description="Whether to include file attachment list")
+
+
+# Response Schemas
+class ReportGenerateResponse(BaseModel):
+    """Response after triggering report generation"""
+    report_id: str = Field(..., description="Unique report identifier")
+    status: ReportStatus = Field(..., description="Initial status (typically 'pending')")
+    message: str = Field(default="Report generation started", description="Status message")
+
+
+class ReportStatusResponse(BaseModel):
+    """Full report metadata response"""
+    report_id: str
+    room_id: str
+    generated_by: str
+    generated_at: datetime
+    status: ReportStatus
+    error_message: Optional[str] = None
+    report_title: Optional[str] = None
+    prompt_tokens: Optional[int] = None
+    completion_tokens: Optional[int] = None
+
+    class Config:
+        from_attributes = True
+
+
+class ReportListItem(BaseModel):
+    """Report item in list response"""
+    report_id: str
+    generated_at: datetime
+    generated_by: str
+    status: ReportStatus
+    report_title: Optional[str] = None
+
+    class Config:
+        from_attributes = True
+
+
+class ReportListResponse(BaseModel):
+    """List of reports for a room"""
+    reports: List[ReportListItem]
+    total: int
+
+
+# AI Report Content Schemas (validated JSON from DIFY)
+class TimelineEvent(BaseModel):
+    """Single event in timeline"""
+    time: str = Field(..., description="Time of event (HH:MM or YYYY-MM-DD HH:MM)")
+    description: str = Field(..., description="Event description")
+
+
+class ParticipantInfo(BaseModel):
+    """Participant information"""
+    name: str = Field(..., description="Participant name")
+    role: str = Field(..., description="Role in incident (e.g., 事件發起人, 維修負責人)")
+
+
+class AIReportContent(BaseModel):
+    """Validated JSON schema from DIFY AI response"""
+    summary: dict = Field(..., description="Event summary with 'content' field")
+    timeline: dict = Field(..., description="Timeline with 'events' list")
+    participants: dict = Field(..., description="Participants with 'members' list")
+    resolution_process: dict = Field(..., description="Resolution process with 'content' field")
+    current_status: dict = Field(..., description="Current status with 'status' and 'description' fields")
+    final_resolution: dict = Field(..., description="Final resolution with 'has_resolution' and 'content' fields")
+
+    @classmethod
+    def validate_structure(cls, data: dict) -> bool:
+        """Validate the basic structure of AI response"""
+        required_keys = ["summary", "timeline", "participants", "resolution_process", "current_status", "final_resolution"]
+        for key in required_keys:
+            if key not in data:
+                return False
+        return True
+
+
+# Error Response
+class ErrorResponse(BaseModel):
+    """Error response"""
+    detail: str