Deprecate batched blas helpers

[ricardoV94]

This should be merged after #1194 to avoid conflicts.

Closes #608

---

📚 Documentation preview 📚: https://pytensor--1215.org.readthedocs.build/en/1215/

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 82.00%. Comparing base (c4fb0cf) to head (7a47d53).
Report is 175 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1215      +/-   ##
==========================================
+ Coverage   81.99%   82.00%   +0.01%     
==========================================
  Files         188      188              
  Lines       48566    48514      -52     
  Branches     8677     8657      -20     
==========================================
- Hits        39823    39786      -37     
+ Misses       6580     6573       -7     
+ Partials     2163     2155       -8     

Files with missing lines	Coverage Δ
pytensor/tensor/blas.py	73.55% <100.00%> (-0.40%)	⬇️
pytensor/tensor/math.py	93.24% <100.00%> (+1.23%)	⬆️
pytensor/tensor/rewriting/blas.py	89.64% <100.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[lucianopaz]

Lgtm except for the test cases that you removed. Why did you have to do that? Maybe it was because there were some redundant tests. If it was because some test stopped working with the changes, then I don’t think that’s right

[ricardoV94]

Lgtm except for the test cases that you removed. Why did you have to do that? Maybe it was because there were some redundant tests. If it was because some test stopped working with the changes, then I don’t think that’s right

No everything works, but those wouldn't be testing batched_dot (the specialized Op) that works with one single batch dimension. They would be testing blockwise matmul or sometimes batched_dot after the rewrite we have if there's no broadcasting needed.