Add more LLM Providers

[rajansagarwal]

Integrate additional LLM providers and ensure reasoning support with new models as well. ie Firecracker, xAI, Vertex... These models need both API key validation and proper integration with a the new models!

[zudsniper]

I think you can deprioritize this, at least below the other issues. Openrouter has cleaned up their act when it comes to things like streaming responses and tool calling, at least in my experience, and openrouter allows you to BYOK within their ecosystem, so they effectively do this for you. In the future though, it would make sense, for sure