Read LLM api endpoint from config #609
Description
Products like LLM Studio expose a server API that is compatible with the OpenAPI format but can actually be powered by virtually any model.
Since cbsh supports OpenAI, it would be nice to be able to configure the endpoint used so I can point it to the LLM Studio endpoint or my company's self-hosted ChatGPT.