Use pyproject.toml, pdm and ruff for improved reproducibility and cleaner code

[nizhib]

@karpathy , thank you for another interesting educational project!

This MR introduces pyproject.toml file to handle project metadata and dependencies in accordance with PEP-621¹.

Using it with pdm² and its lock files is more reliable than an unversioned dependency list stored in requirements.txt.

It also adds a dev dependency for ruff³ linter and code formatter as well as the missing one for pytest.

All the simple issues discovered by ruff and markdownlint⁴ are resolved.

All the tests are still passing as the result of no semantic code changes have been done.

NB: One can also use ruff or nbQA⁵ on Jupyter Notebooks.

TODO: Add more Ruff rule sets like the one for type hints and resolve newly discovered issues.

Footnotes

1. PEP 621 – Storing project metadata in pyproject.toml. ↩

2. PDM is a modern Python package and dependency manager supporting the latest PEP standards. ↩

3. Ruff – An extremely fast Python linter and code formatter, written in Rust. ↩

4. markdownlint – Markdown linting and style checking for Visual Studio Code. ↩

5. nbQA – Run ruff, isort, pyupgrade, mypy, pylint, flake8, black, blacken-docs, and more on Jupyter Notebooks . ↩