Allow interaction with other LLMs

[restlessronin]

Describe the feature or improvement you're requesting

Basic Use Case. I want to use livebook to interact with a local LLM. the easiest way to do this is to use llama.cpp-python as a proxy server that converts openai api to other LLM models. In particular, to begin with, I am only interested in the chat_completion endpoint.

With this proxy, it appears that we only need to change the base url of the api, and we should be able to work with any LLM that is supported by llama.cpp

[restlessronin]

I have tested the new parameterized url and it does seem to work as expected at least for non-streaming completions and chat-completions.