add builtin tool call/return parts for Google file search

Implemented _map_file_search_grounding_metadata following the same pattern
as web search. Extracts retrieval_queries and retrieved_context from
grounding_metadata to create proper BuiltinToolCallPart and
BuiltinToolReturnPart instances.

Author: gorkachea

pydantic_ai_slim/pydantic_ai/models/google.py

@@ -812,6 +812,11 @@ def _process_response_from_parts(
         items.append(web_search_call)
         items.append(web_search_return)
 
+    file_search_call, file_search_return = _map_file_search_grounding_metadata(grounding_metadata, provider_name)
+    if file_search_call and file_search_return:
+        items.append(file_search_call)
+        items.append(file_search_return)
+
     item: ModelResponsePart | None = None
     code_execution_tool_call_id: str | None = None
     for part in parts:
@@ -972,3 +977,28 @@ def _map_grounding_metadata(
         )
     else:
         return None, None
+
+
+def _map_file_search_grounding_metadata(
+    grounding_metadata: GroundingMetadata | None, provider_name: str
+) -> tuple[BuiltinToolCallPart, BuiltinToolReturnPart] | tuple[None, None]:
+    if grounding_metadata and (retrieval_queries := grounding_metadata.retrieval_queries):
+        tool_call_id = _utils.generate_tool_call_id()
+        return (
+            BuiltinToolCallPart(
+                provider_name=provider_name,
+                tool_name=FileSearchTool.kind,
+                tool_call_id=tool_call_id,
+                args={'queries': retrieval_queries},
+            ),
+            BuiltinToolReturnPart(
+                provider_name=provider_name,
+                tool_name=FileSearchTool.kind,
+                tool_call_id=tool_call_id,
+                content=[chunk.retrieved_context.model_dump(mode='json') for chunk in grounding_chunks if chunk.retrieved_context]
+                if (grounding_chunks := grounding_metadata.grounding_chunks)
+                else None,
+            ),
+        )
+    else:
+        return None, None