Add full Ollama provider integration for text generation, code review…

[harshdesaii]

Description

This PR introduces full Ollama provider integration for text generation, code review, embeddings, and image analysis. It enhances the Ensemble CLI with provider-specific routing and automatic provider detection based on available API keys. The CLI now executes all available providers (OpenAI, Ollama, Claude) in parallel, collects their outputs, and intelligently combines them into a single unified response. If any provider fails, it is skipped gracefully without impacting the final result.

Fixes # (issue)

• Default models were updated gpt-4-vision-preview to gpt-4o-mini.
• Added required IMAGE argument that was causing "Missing argument 'IMAGE'" error
• Added _route_specific_provider() for precise provider routing
• Created OllamaProvider adapter with full support for Image analysis, Text generation, Embeddings and Code Review