Feat/litellm

[tqtensor]

What does this PR do?

Integrates LiteLLM and instructor libraries to create a unified interface for multiple vision LLM providers. This simplifies working with different models across providers while maintaining consistent outputs and error handling.

• Addresses issue (#N/A - feature implementation)

Changes implemented

• Replaced direct provider-specific implementations with LiteLLM for standardized model calls
• Added structured output handling with instructor for more reliable response parsing
• Expanded model support to include:
  • OpenAI models (gpt-4o, gpt-4o-mini, gpt-4.1)
  • Google AI Studio models (gemini-1.5-pro, gemini-2.0-flash)
  • DeepSeek models (deepseek-chat)
• Enhanced error handling for provider-specific errors
• Organized prompt templates in a dedicated prompts directory

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Ran make lint and make format to handle lint / formatting issues.
• Ran make test to run relevant tests scripts.
• Read the contributor guidelines.
• Wrote necessary unit or integration tests.