FIX (openai): respect custom API endpoint when creating AsyncOpenAI clients

[toloco]

Respect user-defined OpenAI API endpoints for both Embedding and LLM providers.

Config files defines endpoints here:

• https://github.com/microsoft/NLWeb/blob/main/code/config/config_embedding.yaml#L6
• https://github.com/microsoft/NLWeb/blob/main/code/config/config_llm.yaml#L14

Calls now use the endpoint configured in CONFIG (or OPENAI_ENDPOINT) instead of the hard-coded https://api.openai.com.

• NLWeb already exposes an endpoint field for embedding_provider("openai") and llm_endpoints["openai"], but the value was never forwarded to the AsyncOpenAI client.
• Users relying on Azure OpenAI, local proxies, or gateway URLs had to monkey-patch client creation.

[toloco]

@toloco please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information.

@microsoft-github-policy-service agree [company="{your company}"]

Options:

• (default - no company specified) I have sole ownership of intellectual property rights to my Submissions and I am not making Submissions in the course of work for my employer.

@microsoft-github-policy-service agree

• (when company given) I am making Submissions in the course of work for my employer (or my employer has intellectual property rights in my Submissions by contract or applicable law). I have permission from my employer to make Submissions and enter into this Agreement on behalf of my employer. By signing below, the defined term “You” includes me and my employer.

@microsoft-github-policy-service agree company="Microsoft"

Contributor License Agreement

@microsoft-github-policy-service agree

[Copilot]

Pull Request Overview

This PR ensures that custom OpenAI API endpoints from configuration or environment variables are properly utilized instead of a hard-coded endpoint.

• Added a new method in the llm module to retrieve the API endpoint.
• Updated the AsyncOpenAI client initialization in both llm and embedding modules to use the custom endpoint.
• Adjusted the environment template to align with the expected API endpoint format.

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
code/llm/openai.py	Added get_api_endpoint and updated client initialization to respect custom endpoints.
code/embedding/openai_embedding.py	Added get_openai_api_endpoint and updated client initialization accordingly.
code/.env.template	Updated the OPENAI_ENDPOINT value to a more general base URL.

Comments suppressed due to low confidence (1)

code/embedding/openai_embedding.py:49

• The docstring incorrectly mentions 'API key' instead of 'API endpoint'. Update the docstring to accurately reflect that this function retrieves the API endpoint.

    Retrieve the OpenAI API key from configuration.

[Copilot]

The configured endpoint from provider_config is overwritten unconditionally by the environment variable. Consider checking if the provider_config.endpoint is available before using os.getenv as a fallback.

Suggested change

	# Fallback to environment variable
	endpoint = os.getenv("OPENAI_ENDPOINT")
	if not endpoint:
	error_msg = "OpenAI API endpoint not found in configuration or environment"
	logger.error(error_msg)
	raise ValueError(error_msg)
	# Fallback to environment variable if provider_config.endpoint is not set
	if not endpoint:
	endpoint = os.getenv("OPENAI_ENDPOINT")
	if not endpoint:
	error_msg = "OpenAI API endpoint not found in configuration or environment"
	logger.error(error_msg)
	raise ValueError(error_msg)