Fix stack type hints

[covertg]

Seeking to address #193, making type hints for pt.stack more accurate.

Implementation details

I chose to type hint as TensorLike, defined here as Variable | Sequence[Variable] | npt.ArrayLike. I hope this will address all cases that mypy was complaining about. I also updated the docs and errors accordingly.

Lastly, because there was a case that coerces a list of zero-dim Variables into MakeVector, I also allowed a parallel case of zero-dim numpy arrays to use MakeVector rather than Join and updated a test to cover this case.

Checklist

• Explain motivation and implementation 👆
• Make sure that the pre-commit linting/style checks pass.
  • ^Yes although I had to update my local pre-commit config to isort 5.11.5 due to a bug with pre-commit initialization (error with poetry)
• Link relevant issues, preferably in nice commit messages.
• The commits correspond to relevant logical changes. Note that if they don't, we will rewrite/rebase/squash the git history before merging.
• Are the changes covered by tests and docstrings?
• Fill out the short summary sections 👇

Major / Breaking Changes

• na

New features

• pt.stack([np.array(1), np.array(2)]) will use MakeVector rather than Join

Bugfixes

• More accurate type hints

Documentation

• ...

Maintenance

• ...

Please let me know if you have any feedback!

[codecov-commenter]

Codecov Report

Merging #205 (548e94f) into main (a6e7722) will increase coverage by 0.00%.
The diff coverage is 100.00%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #205   +/-   ##
=======================================
  Coverage   80.06%   80.06%           
=======================================
  Files         170      170           
  Lines       45185    45185           
  Branches     9608     9608           
=======================================
+ Hits        36176    36178    +2     
+ Misses       6794     6793    -1     
+ Partials     2215     2214    -1     

Impacted Files	Coverage Δ
pytensor/tensor/basic.py	90.62% <100.00%> (+0.13%)	⬆️

[michaelosthege]

Thanks, @covertg this is looking good.

When upgrading PyMC to the next PyTensor release we can then revert part of pymc-devs/pymc@e062237