📦 Pull Request Template

Thank you for your contribution! Please complete the checklist and provide relevant details below to help us review your PR effectively.

📋 Summary

This PR adds the implementation of the EGSteal attack into the PyGIP framework.
Specifically, it introduces:

A new EGSteal class under pygip/models/attack/EGSteal/EGSteal.py.

Integration with existing Dataset abstractions (supporting TU datasets such as AIDS, Mutagenicity, NCI1, and NCI109).

Training pipeline for target models (TargetModelGraphClassification) with GIN/GCN/GAT/GraphSAGE backbones.

Explanation alignment logic (CAM-based by default, extendable to GradCAM/GNNExplainer/PGExplainer).

Attack pipeline that queries the target model, generates surrogate data, and trains a surrogate model for evaluation.

This allows PyGIP users to reproduce the results of "How Explanations Leak the Decision Logic: Stealing Graph Neural Networks via Explanation Alignment" and benchmark against other attacks already in the repo.

🧪 Related Issues

✅ Checklist

• My code follows the project's coding style
• I have tested the changes locally on Mutagenicity (baseline run completed successfully)
• I have added necessary documentation (I used the same classes that were already defined)
• I have linked related issues above (if any)
• I have ensured the PR is made from a feature branch (feat/egsteal)

🧠 Additional Context (Optional)