Urban parking demand forecasting P6

[ChiranjeeviVeluri]

This pull request contains the complete implementation of the Urban Parking Demand Forecasting project.

Key additions:

• Graph Neural Network (GCN) model for next time-step parking occupancy prediction
• Spatial graph construction using parking bay proximity
• Feature engineering using hour, day, latitude, and longitude
• Forecasting logic using t -> t+1 occupancy labels
• Area-based filtering for Richmond, Docklands, Melbourne CBD, and other regions
• Interactive visualization with clickable node inspection
• Final predict_by_area.py application for user-driven predictions
• Updated README.md with project overview, file descriptions, setup instructions, and team contributions

Team Contributions:

• Chiranjeevi: Model design, forecasting logic, pipeline integration
• Aditya: Data preparation, model training, hyperparameter tuning
• Aishwarya: Visualization development and UI improvements