[feat] Add Multiple LLM Provider Support (OpenAI/Gemini/Self-Hosted)

[sushen123]

Description
The app currently only supports Ollama for AI features. Allow users to choose between:

• OpenAI
• Google Gemini
• Self-hosted models (custom API endpoints)

Proposed Solution

1. Create an abstract LLM provider interface
2. Add provider selection in settings

I want to implement this! 🙋

Let me know if this aligns with the roadmap! I’d love to collaborate on it.

[sushen123]

hey @khelli07 , Would this be a valid addition to the project? If so, I’d love to contribute and work on it!

[khelli07]

Hey, yeah sure if you are interested with it. Actually it's been a while since I touch this project. But you can see the documentation here.

https://khelli07.medium.com/spending-summer-as-a-google-summer-of-code-contributor-at-rocket-chat-5d194b08df1c

I remember there's still some bugs and limitation on this. So I think you can add the feature you proposed or improve other things. I would suggest you do the thing you proposed first tho. I think it's quite simple since ollama has same API as OpenAI, iirc.

[sushen123]

Thank you for sharing the documentation! I’ll go through it and start working on it

[sushen123]

Hey @khelli07, I’ve made changes to add support for multiple LLM providers, but I’m facing an issue when trying to authorize the scheduler. When I type /schedule authorize, nothing happens. Any idea what might be wrong? Also, should I create a new issue for this, or can I include it when I push the PR?