Fix bug in GP Periodic and WrappedPeriodic kernel full method

[lucianopaz]

Closes #6951

The expand_dims order was wrong for the Periodic and WrappedPeriodic kernels. I added few tests to assert the full method returns the proper axis ordering.

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 👇

Major / Breaking Changes

• ...

New features

• ...

Bugfixes

• Fixed Periodic and WrappedPeriodic covariance kernel full method. A bug was introduced in version 5.7.0 that made full return the transposed covariance matrix.

Documentation

• ...

Maintenance

• ...

---

📚 Documentation preview 📚: https://pymc--6952.org.readthedocs.build/en/6952/

[codecov]

Codecov Report

Merging #6952 (0a628ff) into main (cb64480) will decrease coverage by 0.66%.
Report is 6 commits behind head on main.
The diff coverage is 100.00%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6952      +/-   ##
==========================================
- Coverage   91.29%   90.63%   -0.66%     
==========================================
  Files         100      100              
  Lines       16858    16858              
==========================================
- Hits        15391    15280     -111     
- Misses       1467     1578     +111     

Files	Coverage Δ
pymc/gp/cov.py	98.00% <100.00%> (ø)

... and 9 files with indirect coverage changes

[bwengals]

Reran the failed tests and they passed this time, so, cool