Add GNN code and configurations

[yuvraj333]

📦 Pull Request Template for GNN code and configurations

📋 Summary

This PR integrates the GNN code for ownership verification into the PyGIP framework.
The implementation includes evaluations on the Cora dataset for multiple GNN architectures
(GCN, GAT, GraphSAGE) under both inductive and transductive settings.

The PR includes:

• GNN source code (src/)
• Configuration files (config/)
• Experiments / scripts for running code with run_bengin, run_main and run_robustness
• eval folder and all required updates
• Updated .gitignore to exclude pre-trained models and large result folders

All code is added without including large model files or experimental results to keep the repository clean.

🧪 Related Issues

This PR implements the next step suggested by Bolin Shen (Sep 16, 2025) to integrate ownership verification experiments into the PyGIP framework.

✅ Checklist

• My code follows the project's coding style
• I have tested the changes and verified that they work (runs without model weights or results)
• I have added necessary documentation (README.md, inline code comments)
• I have linked related issues above (if any)
• The PR is made from a feature branch, not main

🧠 Additional Context (Optional)

• Some of the evaluated results and plots are included in experiments/results/. Full code and configurations are added.
• Example scripts are provided in examples/ for reviewers to run the attacks/defenses with minimal setup.
• Large files such as benign_model.pth and result folders (temp_results/, robustness_results/) are excluded via .gitignore.

Some Sample Results are HERE