Probabilty inference for arc transformations

[LukeLB]

Users have asked for the ability to use aditional transformations #6631, also see https://discourse.pymc.io/t/the-sinh-arcsinh-normal-distribution/12136. This ended up being quite fast as the logic is very similar to many of the previous transforms I've implemented.

I've added these transforms to the existing tests, which pass. However, it seems there are a few tests that aren't passing in tests/logprob/test_transform.py. I think they may have been failing prior to me, as they fail on main as well....

...

Checklist

• Explain important implementation details 👆
• Make sure that the pre-commit linting/style checks pass.
• Link relevant issues (preferably in nice commit messages)
• Are the changes covered by tests and docstrings?
• Fill out the short summary sections 👇

New features

• Arcsinh transform
• Arccosh transform
• Arctanh transform

---

📚 Documentation preview 📚: https://pymc--6775.org.readthedocs.build/en/6775/

[codecov]

Codecov Report

Merging #6775 (5dd3b02) into main (154f5b0) will decrease coverage by 0.69%.
The diff coverage is 86.95%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6775      +/-   ##
==========================================
- Coverage   91.89%   91.21%   -0.69%     
==========================================
  Files          95       95              
  Lines       16147    16168      +21     
==========================================
- Hits        14839    14747      -92     
- Misses       1308     1421     +113     

Impacted Files	Coverage Δ
pymc/logprob/transforms.py	94.24% <86.95%> (-0.36%)	⬇️

... and 4 files with indirect coverage changes

[ricardoV94]

@LukeLB which tests failed exactly? There were some jobs that for some reason were cancelled and didn't run. I restarted them manually now.

[LukeLB]

Faiures are below:
FAILED test_transforms.py::test_reciprocal_rv_transform - AttributeError: module 'pytensor.tensor' has no attribute 'get_underlying_scalar_constant_value'
FAILED test_transforms.py::test_sqr_transform - AttributeError: module 'pytensor.tensor' has no attribute 'get_underlying_scalar_constant_value'
FAILED test_transforms.py::test_negative_value_odd_power_transform - AttributeError: module 'pytensor.tensor' has no attribute 'get_underlying_scalar_constant_value'
FAILED test_transforms.py::test_scan_transform - AttributeError: 'NoneType' object has no attribute 'inputs'

EDIT: Ah this will be an envrironment thing I bet! Will try with new env
EDIT2: Yep that was it, all tests pass locally. I should learn to update my environment more frequently!

[ricardoV94]

Let me know if it works and if you think we should merge

[ricardoV94]

Thanks @LukeLB!