Update LLM provider support (Adding Claude integration and deprecate Gemini models (2.0 Flash and 2.0 Flash Lite)

[benevolentbandwidth-nurtekin]

This issue involves 2 tasks to be completed:

1. We have to add Claude models as a backend provider.
2. Replace deprecated Gemini 2.0 models (Gemini 2.0 Flash and Gemini 2.0 Flash Lite).