Build wheels for the fork of scikit-learn here

[adam2392]

We want a CI pipeline (and/or local pipeline) that can build wheels for:

• windows
• Linux
• Mac Intel (x64)
• Mac M1 (Arm)

and we can attach those wheels to a specific release we make (these are called "nightly wheels" in scipy/pytorch/tensorflow/etc.). Then we can pip install directly from those wheels. Use https://cibuildwheel.readthedocs.io/en/stable/.

Then create a stable release that has all these wheels pip installable

Then have sktree rely on these wheels for v0.1

I added more notes on the call, feel free to add them here for documentation @jshinm.

[adam2392]

Setup Anaconda space and then fix up documentation to describe where the wheels are found:

https://scikit-learn.org/stable/developers/advanced_installation.html

Perhaps overall, the fork, which we can call scikit-learn-tree would just scrap most of the docs folder and create a separate doc folder for documentation of the fork, which really just requires two things:

1. how to install
2. documentation on changes wrt sklearn main
3. link to sklearn documentation for anything else

[adam2392]

@jovo and @sampan501 the latest wheels are released under v1.2.2 and under pypi scikit-learn-tree, which is basically an alias and stand-in for scikit-learnv1.2.2, but with the adapted changes underneath the hood.

Any new release process for us more or less follows now the process of scikit-learn upstream, but we just have to manually upload the wheels ourselves (they are built on GH actions and Cirrus CI).

[adam2392]

Closed by #41