Remove t suffix from Model methods

[cuchoi]

Closes #5859

To-do:

• Rename model.logpt, model.dlogpt and model.d2logpt to pymc.model.Model without the t suffix. Keep the old ones as a simple wrapper that issues a FutureWarning in between
• Update model.datalogpt, model.varlogpt, model.observedlogpt, and model.potentiallogpt, distributions.joint_logpt as well
• Change all the tests to make use of the new names
• Add some tiny tests to make sure the deprecation is being issued
• Update documentation docs/source/contributing/developer_guide.rst and docs/source/learn/core_notebooks/pymc_aesara.ipynb

[codecov]

Codecov Report

Merging #5863 (a33cd0b) into main (a5f3e45) will increase coverage by 0.04%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #5863      +/-   ##
==========================================
+ Coverage   89.45%   89.50%   +0.04%     
==========================================
  Files          73       73              
  Lines       13234    13267      +33     
==========================================
+ Hits        11839    11874      +35     
+ Misses       1395     1393       -2     

Impacted Files	Coverage Δ
pymc/backends/arviz.py	90.28% <ø> (ø)
pymc/distributions/__init__.py	100.00% <ø> (ø)
pymc/distributions/continuous.py	97.98% <ø> (ø)
pymc/smc/smc.py	96.45% <ø> (ø)
pymc/distributions/logprob.py	97.72% <100.00%> (+0.07%)	⬆️
pymc/model.py	87.87% <100.00%> (+0.61%)	⬆️
pymc/sampling.py	82.41% <100.00%> (ø)
pymc/sampling_jax.py	96.93% <100.00%> (ø)
pymc/step_methods/metropolis.py	83.36% <100.00%> (ø)
pymc/step_methods/mlda.py	96.37% <100.00%> (ø)
... and 3 more

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[cuchoi]

Should I add/replicate test for the t ending functions? I only added FutureWarning tests for those but no functionality tests (for example if there is a typo and dlogpt calls logpt -- I double checked but you never know :))

[ricardoV94]

Should I add/replicate test for the t ending functions? I only added FutureWarning tests for those but no functionality tests (for example if there is a type and dlogpt and calls logpt -- I double checked but you never know :))

If you want you can try to monkey patch the methods to make sure what was called internally, but otherwise there's no need. I'll review carefully with some fresh eyes to confirm it looks alright ;)

[michaelosthege]

This looks like something we should fast-track into a 4.0.1 release to minimize the friction for people migrating their logp* calls.

@ricardoV94 can you review it?

[ricardoV94]

I'll review it on Monday

[ricardoV94]

Looks great, left some suggestions and highlighted one missed property with t suffix.

[ricardoV94]

We shouldn't change the old Release Notes as this refer to the already released V4.0.0. Instead you should add a new release note in the most recent section explaining the changes from t suffix to no suffix.

[michaelosthege]

This weekend I already created a 4.0.1 section in the release notes.
You'll have to rebase, otherwise there will be git conflicts..

[ricardoV94]

Looks great 👍

[cuchoi]

I rebased but now some changes appear done not by me appear as they were new, is that ok? Do you know how to fix it? Fixed

Also, should I add this change to the Release Notes of 4.0.1? Added release notes

[twiecki]

Thanks @cuchoi!