feat: 新增风险规则生成引擎与知识图谱可视化

后端新增风险规则自动生成和模板执行服务，支持从规则资产
批量生成并持久化风险规则文件；知识库入库日志增强图谱
查询和本地 RAG 回退，前端审计页面增加风险规则模型和流
程图组件，知识入库面板拆分为图谱可视化子组件，报销创
建页面增加引导式流程模型，更新知识库索引数据。

server/src/app/services/knowledge_rag.py

@@ -17,6 +17,7 @@ from app.services.knowledge_ingest_log import (
     build_ingest_document_summary,
     build_ingest_status_summary,
 )
+from app.services.knowledge_rag_local import query_local_text_chunks
 from app.services.knowledge_rag_runtime import (
     KnowledgeRagError,
     RuntimeModelConfig,
@@ -95,6 +96,37 @@ class KnowledgeRagService:
                 "message": "请先输入要检索的知识库问题。",
             }
 
+        workspace = (
+            os.environ.get("LIGHTRAG_WORKSPACE", DEFAULT_LIGHTRAG_WORKSPACE).strip()
+            or DEFAULT_LIGHTRAG_WORKSPACE
+        )
+        local_result = query_local_text_chunks(
+            lightrag_root=(self.storage_root / "knowledge" / ".lightrag").resolve(),
+            workspace=workspace,
+            query=normalized_query,
+            limit=limit,
+        )
+        if local_result.confident:
+            return {
+                "result_type": "knowledge_search",
+                "query": normalized_query,
+                "record_count": len(local_result.hits),
+                "hits": local_result.hits,
+                "references": [
+                    str(item.get("code") or "").strip()
+                    for item in local_result.hits
+                    if str(item.get("code") or "").strip()
+                ],
+                "raw_references": [],
+                "metadata": {
+                    "retrieval_strategy": "local_text_chunks",
+                    "elapsed_seconds": round(local_result.elapsed_seconds, 4),
+                    "total_chunks": local_result.total_chunks,
+                    "best_score": local_result.best_score,
+                },
+                "message": f"已从本地知识块中检索到 {len(local_result.hits)} 条相关内容。",
+            }
+
         try:
             runtime = self._get_runtime()
             raw = runtime.query_data(normalized_query, conversation_history=conversation_history)