feat: Add Deduplication Agent with embedding and LLM methods

Implement a new Deduplication Agent that identifies and groups similar
transformation descriptions. Supports two deduplication methods:
- Embedding: Fast vector similarity comparison using cosine similarity
- LLM: Accurate pairwise semantic comparison (slower but more precise)

Backend changes:
- Add deduplication router with /deduplicate endpoint
- Add embedding_service for vector-based similarity
- Add llm_deduplication_service for LLM-based comparison
- Improve expert_transformation error handling and progress reporting

Frontend changes:
- Add DeduplicationPanel with interactive group visualization
- Add useDeduplication hook for state management
- Integrate deduplication tab in main App
- Add threshold slider and method selector in sidebar

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

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

backend/app/models/schemas.py

@@ -232,3 +232,38 @@ class ExpertTransformationRequest(BaseModel):
     # LLM parameters
     model: Optional[str] = None
     temperature: Optional[float] = 0.7
+
+
+# ===== Deduplication Agent schemas =====
+
+class DeduplicationMethod(str, Enum):
+    """去重方法"""
+    EMBEDDING = "embedding"  # 向量相似度
+    LLM = "llm"              # LLM 成對判斷
+
+
+class DeduplicationRequest(BaseModel):
+    """去重請求"""
+    descriptions: List[ExpertTransformationDescription]
+    method: DeduplicationMethod = DeduplicationMethod.EMBEDDING  # 去重方法
+    similarity_threshold: float = 0.85  # 餘弦相似度閾值 (0.0-1.0)，僅 Embedding 使用
+    model: Optional[str] = None  # Embedding/LLM 模型
+
+
+class DescriptionGroup(BaseModel):
+    """相似描述分組"""
+    group_id: str                          # "group-0", "group-1"...
+    representative: ExpertTransformationDescription  # 代表描述
+    duplicates: List[ExpertTransformationDescription]  # 相似描述
+    similarity_scores: List[float]         # 每個重複項的相似度分數
+
+
+class DeduplicationResult(BaseModel):
+    """去重結果"""
+    total_input: int           # 輸入描述總數
+    total_groups: int          # 分組數量
+    total_duplicates: int      # 重複項數量
+    groups: List[DescriptionGroup]
+    threshold_used: float
+    method_used: DeduplicationMethod       # 使用的去重方法
+    model_used: str                        # 使用的模型