fixed reasoning

backend/app/llm/client.py

@@ -147,8 +147,9 @@ class LLMClient:
         *,
         temperature: float = 0.7,
         model: str | None = None,
+        for_extraction: bool = False,
     ) -> dict[str, Any]:
-        use_tools = bool(tools) and self.tools_enabled
+        use_tools = bool(tools) and self.tools_enabled and not for_extraction
         kwargs: dict[str, Any] = {
             "model": model or self.model,
             "messages": messages,
@@ -156,9 +157,12 @@ class LLMClient:
         }
         if use_tools:
             kwargs["tools"] = tools
-        extra_body = self._reasoning_extra_body()
-        if extra_body:
-            kwargs["extra_body"] = extra_body
+        if for_extraction:
+            kwargs["extra_body"] = {"reasoning": {"effort": "none"}}
+        else:
+            extra_body = self._reasoning_extra_body()
+            if extra_body:
+                kwargs["extra_body"] = extra_body
 
         response = await self.client.chat.completions.create(**kwargs)
         message = response.choices[0].message