[WIP] Add pyproject.toml and CI workflow configuration

[Copilot]

Thanks for asking me to work on this. I will get started on it and keep this PR's description up to date as I form a plan and make progress.

Original prompt

AstroVision is a deep learning pipeline for morphological galaxy classification
using PyTorch. The project is already structured as follows:

src/astrovision/

• data_loader.py (Galaxy10 DECaLS HDF5 loader, PyTorch Dataset)
• models.py (SimpleCNN from scratch + EfficientNet-B0 fine-tuning)
• trainer.py (training loop with W&B tracking)
• gradcam.py (Grad-CAM visualization)
  notebooks/
• 01_galaxy10_cnn.ipynb
  requirements.txt (torch, torchvision, h5py, wandb, scikit-learn, matplotlib)

DO NOT regenerate these files. Instead, please generate only:

1. pyproject.toml with project metadata (name=astrovision, python>=3.11,
   all deps from requirements.txt, setuptools build backend)

2. .github/workflows/ci.yml — GitHub Actions workflow that:

   • Triggers on push to main and pull requests
   • Runs on ubuntu-latest with Python 3.11
   • Installs dependencies with pip
   • Runs pytest on tests/ directory
   • Lints with ruff

3. tests/test_models.py — pytest unit tests that verify:

   • SimpleCNN forward pass produces correct output shape (B, 10)
   • EfficientNetGalaxy forward pass produces correct output shape (B, 10)
   • build_model factory raises ValueError on unknown model name
   • GradCAM instantiates without errors

4. tests/test_data_loader.py — pytest unit tests that:

   • Mock the HDF5 file and verify Galaxy10Dataset len and getitem
   • Verify get_transforms returns correct Compose objects for train/val

5. ruff.toml — ruff linter config (line-length=100, target Python 3.11)

[Copilot]

Copilot wasn't able to review any files in this pull request.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.