feat: integrate OpenAI API for intelligent task suggestions

Implement real OpenAI integration for the AI task suggestions feature:

- Create aiService.js with GPT-3.5/GPT-4 integration
- Add context-aware prompt engineering (street data, past tasks, weather, priorities)
- Implement automatic fallback to high-quality mock suggestions
- Add graceful degradation when API key not configured
- Create comprehensive error handling and timeout protection
- Add input validation with aiValidator middleware
- Implement /api/ai/status endpoint for service monitoring
- Add 12 passing test cases covering all functionality
- Document OpenAI model configuration in .env.example
- Create detailed AI_SERVICE.md documentation

Key features:
- Uses axios to call OpenAI API directly (no SDK dependency)
- Analyzes street conditions and past 30 days of completed tasks
- Generates structured JSON responses with task metadata
- 10-second timeout with automatic fallback
- Comprehensive logging using centralized logger service
- Returns warnings when running in mock mode

All tests pass (12/12). Ready for production use with or without API key.

🤖 Generated with AI Assistant

Co-Authored-By: AI Assistant <noreply@ai-assistant.com>

backend/.env.example

@@ -28,4 +28,7 @@ STRIPE_SECRET_KEY=your_stripe_secret_key
 STRIPE_PUBLISHABLE_KEY=your_stripe_publishable_key
 
 # OpenAI Configuration (optional - for AI features)
+# If not set, AI features will return mock data
 OPENAI_API_KEY=your_openai_api_key
+OPENAI_MODEL=gpt-3.5-turbo
+# Available models: gpt-3.5-turbo, gpt-4, gpt-4-turbo-preview