Fix warn_treedepth looking at the wrong stat

[michaelosthege]

Fixes a bug in warn_treedepth that was introduced in code refactoring PR #6192.

Closes #6587

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 👇

Bugfixes

• Fixes incorrect warnings about reaching max treedepth.

[codecov]

Codecov Report

Merging #6591 (8297c5b) into main (cf6d4ce) will increase coverage by 0.01%.
The diff coverage is 100.00%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6591      +/-   ##
==========================================
+ Coverage   92.03%   92.04%   +0.01%     
==========================================
  Files          92       92              
  Lines       15535    15535              
==========================================
+ Hits        14297    14299       +2     
+ Misses       1238     1236       -2     

Impacted Files	Coverage Δ
pymc/distributions/discrete.py	99.22% <ø> (ø)
pymc/model.py	89.52% <ø> (ø)
pymc/stats/convergence.py	94.73% <100.00%> (+4.38%)	⬆️

... and 1 file with indirect coverage changes

[ricardoV94]

Can we add a parametrize to whatever test checks this warning, that it is not issued when it shouldn't? Maybe just use the model from the original issue