Wrap sigma parameter with as_tensor_variable() before get_tau_sigma()

[enessinanparildi]

For the cases where the sigma parameter is a vector of random variables instead of fixed values, in get_tau_sigma
if np.any(sigma_ <= 0) throws the following error: TypeError: Variables do not support boolean operations.
Sigma parameter normally is wrapped by as_tensor_variable() after get_tau_sigma. Moving the line sigma= pt.as_tensor_variable(sigma) before get_tau_sigma() resolves this issue. I ran some tests by supplying sigma argument with different types of values. They are all fine.

Bugfixes

• For the cases where the sigma parameter is a vector of random variables instead of fixed values, in get_tau_sigma
  if np.any(sigma_ <= 0) throws the following error: TypeError: Variables do not support boolean operations.

---

📚 Documentation preview 📚: https://pymc--6745.org.readthedocs.build/en/6745/

[welcome]

💖 Thanks for opening this pull request! 💖 The PyMC community really appreciates your time and effort to contribute to the project. Please make sure you have read our Contributing Guidelines and filled in our pull request template to the best of your ability.

[ricardoV94]

That will avoid the negative check every time. I think we can remove the check because I don't think users would ever pass non symbolic negative sigmas.

If they do, the new model.debug should quickly find the problem anyway.

[ricardoV94]

Closing in favor of #6988