Add base_url param support for LLM / VLM

[d6l8]

Is your feature request related to a problem? Please describe.
I need to use OpenAI models with the base_url param, but simply putting OPENAI_API_KEY in the .env file doesn't work because the function create_chat_completion in the gpt-researcher\gpt_researcher\utils\llm.py doesn't support this param. And your code for the OpenAI embedding model supports this param.

_embeddings = OpenAIEmbeddings(
    model=model,
    openai_api_key=os.getenv("OPENAI_API_KEY", "custom"),
    openai_api_base=os.getenv(
        "OPENAI_BASE_URL", "http://localhost:1234/v1"
    ),  # default for lmstudio

Describe the solution you'd like
Add support for the base_url param in the func create_chat_completion, which reads the base_url param from the .env file like above.