Dimitri Papadopoulos Orfanos

The CI error looks like a transient GitHub error: ``` [1/9] Performing download step (download, verify and extract) for 'vendored_library-populate' -- Downloading... dst='/tmp/tmpgmgqqevi/build/_deps/vendored_library-subbuild/vendored_library-populate-prefix/src/0cdc758a56360bf58a851fe91085a327ec97685a.tar.gz' timeout='none' inactivity timeout='none' -- Using src='https://github.com/bitcoin-core/secp256k1/archive//0cdc758a56360bf58a851fe91085a327ec97685a.tar.gz' CMake...

It looks like `hatch fmt --check` runs the benchmarks, and for that it needs to build fastecdsa.

It's indeed the benchmark failing while trying to build `fastecdsa==3.0.1`: https://github.com/ofek/coincurve/actions/runs/18672653134/job/53236594428#step:11:40 I suspect the `env` might be shared by all steps. I will nevertheless move `env` to the proper step....

> Also need to update the tox environments: Done. @effigies It's now ready for reviewing.

Bumping ruff from 0.1.9 to 0.3.0 is unexpectedly intrusive - lots of change by `ruff format`.

I think I have addressed all the issues - plus some minor new changes.

I think I have addressed all of them. Note that I rebased to fix a merge conflict.

Or not :smile: The quantity of changes pushes the GitHub web interface to its limits. I have a hard time navigating the comments. Still one to address, almost there.

The linter target should be `py38` but `hatch fmt --check` clearly uses a more recent target, hence the CI errors. Any clue how the target Python version is chosen?

I have extracted some commits to #2063 to simplify this PR.