Make it easier to run studies from git checkouts

[mikix]

• When loading a builder, we add the study dir's parent folder into sys.path, so that builders can import helper code. This is basically equivalent to doing PYTHONPATH=. from a checkout.
• When searching for the study dir, and we find a manifest.toml, actually read the manifest to find the study_prefix instead of using the directory name for the study name. This allows more natural python package layouts (rather than module_name/study_name, you can just do module_name)

This also bumps ruff's version to 0.4.4 and has the commit hook just auto-commit any formatting changes.

Checklist

• [x] Consider if documentation (like in docs/) needs to be updated
• [x] Consider if tests should be added
• [x] Update template repo if there are changes to study configuration

[mikix]

I didn't see any changes needed in the template repo, mostly because (1) it doesn't include any builders that try to import anything, (2) it doesn't use the two-level directory setup that some released studies like covid have had to use, and (3) its instructions talk about renaming the study folder already to match a python module name and to also rename study_prefix.

So I think it should be fine both before and after this change, but your naming can be more flexible after this PR

[dogversioning]

I didn't see any changes needed in the template repo, mostly because (1) it doesn't include any builders that try to import anything, (2) it doesn't use the two-level directory setup that some released studies like covid have had to use, and (3) its instructions talk about renaming the study folder already to match a python module name and to also rename study_prefix.

So I think it should be fine both before and after this change, but your naming can be more flexible after this PR

We should maybe chat about whether maintaining the template repo even makes sense - The example config file in the docs might be the 'right' way to telegraph how to make studies at this point.