feat: drop Python 3.7, pre-commit.ci autoupdate

[pre-commit-ci]

updates:

• github.com/asottile/setup-cfg-fmt: v2.2.0 → v2.4.0
• github.com/asottile/pyupgrade: v3.4.0 → v3.10.1
• github.com/psf/black: 23.3.0 → 23.7.0
• github.com/PyCQA/flake8: 6.0.0 → 6.1.0
• github.com/pre-commit/mirrors-mypy: v1.2.0 → v1.5.0
• github.com/codespell-project/codespell: v2.2.4 → v2.2.5

[jonas-eschle]

This upgrades to Python >= 3.8, do we have any plans to move mplhep further and remove 3.7 support? Same goes for mplhep_data, the dep (scikit-hep/mplhep_data#40)

[matthewfeickert]

do we have any plans to move mplhep further and remove 3.7 support

Python 3.7 is EOL as of 2023-06-27 so people should start dropping it. 👍

[andrzejnovak]

Do we need to drop this right now? @matthewfeickert I don't know what the best practices are here now, but naively I would say that if we don't have any compatibility issues there's no reason to force it?

[jonas-eschle]

While true, I would in general try to go with the flow: upgrading from 3.6 on isn't the same as doing a 2 to 3 upgrade and will keep people motivated to move for the newest features. Keep in mind that it will hold back the mplhep code as it won't have new Python features (typing, walrus operator,...). I would suggest to roughly follow NEP 29 guidelines as also mentioned in Scikit-HEP, with possibly the best argument: "we don't need to support anything that Numpy doesn't" -> Python 3.7 users will still get mplhep, just not the very newest version