feat: Add user defined openai model provider (#124)

[MiaoDX]

Change-Id: Ice3c42381021e474d751528a0b5c4726998a6acb

Hi,

I am using other OpenAI model providers, with custom base_url and model. So, it should be nice to be configurable and won't hurt others.

And, inside llm_api.py and web_scraper.py, change to use python3 directly, to be consistent with other tools.

[Copilot]

Copilot reviewed 2 out of 3 changed files in this pull request and generated no comments.

Files not reviewed (1)

• .env.example: Language not supported

Comments suppressed due to low confidence (2)

tools/llm_api.py:71

• Consider validating or normalizing the base_url to ensure it adheres to the expected URL format. This can help avoid runtime issues with misconfigured endpoints.

base_url = os.getenv('OPENAI_BASE_URL', "https://api.openai.com/v1")

tools/llm_api.py:145

• Review the default model fallback to ensure 'gpt-4o' matches the expected model identifier for the provider. Confirm that this behavior is aligned with the overall configuration strategy.

model = os.getenv('OPENAI_MODEL_DEPLOYMENT', 'gpt-4o')

[grapeot]

Thanks for the PR!