[FEAT] Wire End-to-End Token Streaming from Model APIs to Frontend

[shrixtacy]

Description

The WebSocket endpoint at /ws processes requests synchronously — it calls ai_council.process_request(), waits for the complete response, then sends the entire result as one JSON payload. No token-by-token streaming from model APIs to the frontend.

In BEST_QUALITY mode, the user sees a blank screen for the entire multi-model execution duration.

Component

web_app/backend/main.py:332-378, ai_council/execution/agent.py

Acceptance Criteria

In BEST_QUALITY mode, the user sees partial responses appearing progressively rather than a blank screen followed by a complete answer. Each model's response streams as tokens arrive.

[github-actions]

🙏 Thank you for raising this issue!

We'll review it as soon as possible. We truly appreciate your contributions! ✨

📚 Before you start working:

• ✅ Make sure you've read the README.md
• ✅ Review CONTRIBUTING.md for guidelines
• ✅ Familiarize yourself with CODE_OF_CONDUCT.md

⚠️ Important: Please do NOT create a PR until this issue has been assigned to you.

💡 Want to work on this? Comment "assign me" or "I want to work on this" to get assigned!