Feature(LiteLLM): Wire LiteLLM Docker profile with application config overrides (provider and embedder) - #533

Summary

This PR is part of a 4-PR integration effort to add LiteLLM support to DeepWiki-Open while maintaining full backward compatibility.

PR Series

• New Feature: Add optional LiteLLM Docker Compose setup for local LLM gateway #526 — Add optional LiteLLM Docker Compose setup for local LLM gateway
• Feature: Introduce LiteLLM client for multi-provider model routing #529 — Introduce LiteLLM client for multi-provider model routing
• This PR — Wire Docker Compose with LiteLLM provider and configuration layer
• Next — Documentation and usage guide updates

Summary (this PR)

This PR completes the deployment-layer integration for LiteLLM by wiring the Docker Compose setup to application-level provider and embedding configurations.

It ensures that when LiteLLM is enabled via Docker, the correct provider configuration and model registry are automatically applied.

🔧 Changes

Configuration Layer

• Added generator.litellm.json with full provider registry for:
  • OpenAI, OpenRouter, Dashscope, Google, Azure, Bedrock, Ollama, LiteLLM
• Added embedder.litellm.json for LiteLLM-based embedding configuration
• Ensured LiteLLM is the default provider when selected

Docker Integration

• Mounted LiteLLM-specific config overrides in docker-compose-litellm.yml
  • generator.litellm.json → generator.json
  • embedder.litellm.json → embedder.json

Provider Alignment

• Ensured consistency between:
  • Provider routing (PR2)
  • Docker deployment (PR1)
  • Configuration system (this PR)

🧠 Design Notes

• LiteLLM remains fully optional and opt-in
• No changes to default runtime behavior unless explicitly enabled
• All existing providers (OpenAI, OpenRouter, AWS, Google, etc.) remain unchanged
• Configuration overrides are isolated to LiteLLM Docker profile only

🔄 Compatibility

This change is fully backward compatible:

• Existing deployments are unaffected
• Default provider behavior remains unchanged
• LiteLLM is only activated via Docker profile selection
• No breaking API or configuration changes

🧪 Testing

Tested with:

• Docker Compose LiteLLM profile activation
• Provider selection switching (OpenAI ↔ LiteLLM)
• Embedding pipeline with LiteLLM embedder config
• Multi-provider model routing behavior
• End-to-end WebSocket query flow

📝 Notes for Reviewers

This PR focuses strictly on deployment + configuration wiring.

• PR Feature: Introduce LiteLLM client for multi-provider model routing #529 introduced the provider abstraction
• PR New Feature: Add optional LiteLLM Docker Compose setup for local LLM gateway #526 introduced the Docker gateway
• This PR connects both via config + compose overrides

A documentation PR will follow to explain usage and setup flows.