Bug: App ignores OPENAI_BASE_URL overrides and fails to route to custom endpoints (Groq)

Description: I am trying to use Groq as an OpenAI-compatible endpoint with the llama-3.3-70b-versatile model. The application ignores environment variable overrides for the base URL and continues routing requests to OpenAI's default servers. This results in a 404 error because OpenAI does not recognize the custom model name.

Steps to reproduce / Troubleshooting applied:

1. Updated ~/.hedge-fund/.env and local .env files with OPENAI_BASE_URL and OPENAI_API_BASE pointing to https://api.groq.com/openai/v1.
2. Set the environment variable HEDGE_FUND_LLM_MODEL=llama-3.3-70b-versatile.
3. Forced the environment variables directly via Windows CMD (set command) before running the application.
4. Attempted to modify client.py locally to explicitly pass the base_url parameter to the ChatOpenAI constructor.

None of these steps prevented the app from hitting the default OpenAI servers.

Error message received: LLM call failed: Error code: 404 - {'error': {'message': 'The model llama-3.3-70b-versatile does not exist or you do not have access to it.', 'type': 'invalid_request_error', 'code': 'model_not_found'}}