LiteLLM support and reference config

[rcarmo]

Is your feature request related to a problem? Please describe.
I would like to be able to use LiteLLM instead of direct OpenAI or Ollama configs

Describe the solution you'd like
I would like to have the ability to easily set a LiteLLM endpoint and key and pick the models to use.

Describe alternatives you've considered
Hacking this myself overriding the OpenAI endpoint config (which is confusing since the app cannot be solely configured via environment variables and things are inconsistent)

[satonotdead]

I like it, should enable openrouter and and more options as well.