Improve workflows for habits 4-7 with GitHub link extraction and MCP evaluation focus

[Copilot]

This PR addresses the requirements to improve workflows for habits 4, 5, 6, and 7 by implementing three key enhancements:

🔗 Actual GitHub Links in Generated Reports

Previously, workflows generated reports with generic references but didn't include actual GitHub URLs from MCP tool outputs. This PR introduces a new framework/github_utils.py module that:

• Extracts GitHub URLs directly from MCP tool responses and AI messages
• Formats links with meaningful descriptions (e.g., "microsoft/vscode - Issue #123")
• Automatically includes actual repository links in all generated markdown reports

# Example of improved link extraction
links = extract_github_links_from_messages(ai_messages)
# Returns: ['https://github.com/microsoft/vscode/issues/123', ...]

markdown = format_github_links_for_markdown(links, "MCP Resources")
# Generates properly formatted markdown with actual GitHub links

⚡ Minimized Tool Usage with Clear Separation

The original workflow had unclear boundaries between data collection and summarization. This PR restructures each workflow with distinct phases:

• Data Collection Node: Focused searches with minimal tool complexity
• Tools Node: Dedicated GitHub MCP interactions
• Link Extraction Node: Processes tool responses to extract GitHub URLs
• Synthesis Node: Creates rich reports with embedded actual links

This clear separation ensures each step handles minimal tools and has a single focused responsibility.

🎯 MCP Evaluation Framework Focus for Habit 4

Habit 4 now specifically targets actionable issues and work related to building evaluation frameworks for MCP that help AI optimize software for companies:

# Habit 4 now searches for:
- "MCP evaluation framework" OR "model context protocol testing"
- Issues labeled "evaluation", "framework", "testing", "collaboration"
- Repositories with "mcp" OR "model-context-protocol" topics
- Collaborative development patterns in AI tooling

📁 Individual Workflow Separation

The monolithic habit4567_summary workflow has been split into focused individual workflows:

• habit4-win-win: MCP evaluation framework and collaborative development
• habit5-understand: Understanding-first communication and review analysis
• habit6-synergize: Multi-tool integration and synergistic collaboration
• habit7-sharpen: Learning opportunities and skill development

Each workflow maintains domain-specific search strategies and generates targeted reports with actual GitHub links extracted from MCP responses.

🔧 Additional Improvements

• Robust Prompt Management: Fixed framework/prompt_manager.py to handle missing Langfuse configuration gracefully
• Coordinator Workflow: Updated original habit4567-summary to coordinate individual workflows
• Comprehensive Testing: All workflows build successfully and GitHub utilities are validated

The improved workflows now generate actionable reports with real GitHub repository links, clear separation of concerns, and specific focus on MCP evaluation opportunities for AI optimization.

---

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