[WIP] Fix issue by removing Gemini for Perplexity API

[Copilot]

• Investigate current LLM integration (lib/llm.ts)
• Confirm Gemini API code is still present in lib/llm.ts
• Search for existing Perplexity API implementation (not found)
• Reply to comment requesting Perplexity implementation details
• BLOCKED: Awaiting response with Perplexity API details
• Update lib/llm.ts to use Perplexity instead of Gemini
• Remove @google/generative-ai from package.json
• Update .env.example to reference PERPLEXITY_API_KEY
• Test chat functionality with new integration
• Verify no remaining Gemini references

Current Status:
Waiting for @DHCross to provide:

1. Perplexity API endpoint and auth method
2. Whether to use existing integration or create new one
3. API interface preferences (streaming, etc.)

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.