[Feature] Integration with Litellm To get 100+ Models easily

[Greatz08]

These Many LLM's might not be enough for many of us

OpenAI (ChatGPT)
Azure OpenAI
Claude
Google Gemini Pro
Ollama (enable access to local models like llama2, Mistral, Mixtral, codellama, vicuna, yi, and solar)
ChatGLM-6B

There is one great solution Litellm to fix this issue universally and make things easy for devs and for people too in my opinion. People can easily use many more models easily just need to run one docker container and maintain small extremely easy .yaml config for models and thats all and in interface and in backend people need it to setup custom api server url through which we can route through litellm to connect easily with many other models and this is how openwebui like great projects have implemented too so that people can use many more models easily.