fix: apply base_url to GoogleClient to make v1beta endpoint work

The previous change in cli/utils.py set base_url to v1beta, but it was
never passed to ChatGoogleGenerativeAI. This fix ensures the custom
endpoint is actually used, resolving the 404 error for embeddings.

Addresses Gemini Code Assist review feedback on PR #385.

tradingagents/llm_clients/google_client.py

@@ -38,6 +38,10 @@ class GoogleClient(BaseLLMClient):
         """Return configured ChatGoogleGenerativeAI instance."""
         llm_kwargs = {"model": self.model}
 
+        # Support custom base_url (e.g., v1beta endpoint for embeddings)
+        if self.base_url:
+            llm_kwargs["base_url"] = self.base_url
+
         for key in ("timeout", "max_retries", "google_api_key", "callbacks", "http_client", "http_async_client"):
             if key in self.kwargs:
                 llm_kwargs[key] = self.kwargs[key]