BUG: Suppress numpy warnings in PyTensor Elemwise perform

[asifzubair]

Describe the issue:

This is not really a bug, but a request to suppress numpy warnings when doing some valid operations. These warnings came up when working on the CensovedRV. @ricardoV94 provided some context for them:

We used to constant-fold in C which doesn't have such warnings, and now we do in Python.

Reproduceable code example:

import numpy as np
import pymc as pm
from pymc import logcdf
norm = pm.Normal.dist(0, 1)
logcdf(norm, np.inf).eval()
logcdf(norm, -np.inf).eval()

Warning message:

>>> logcdf(norm, np.inf).eval()
/opt/conda/envs/pymc-dev/lib/python3.13/site-packages/pytensor/tensor/elemwise.py:710: RuntimeWarning: invalid value encountered in subtract
  variables = ufunc(*ufunc_args, **ufunc_kwargs)
array(-0.)
>>> logcdf(norm, -np.inf).eval()
/opt/conda/envs/pymc-dev/lib/python3.13/site-packages/pytensor/tensor/elemwise.py:710: RuntimeWarning: divide by zero encountered in log1p
  variables = ufunc(*ufunc_args, **ufunc_kwargs)
/opt/conda/envs/pymc-dev/lib/python3.13/site-packages/pytensor/tensor/elemwise.py:710: RuntimeWarning: divide by zero encountered in log
  variables = ufunc(*ufunc_args, **ufunc_kwargs)
array(-inf)

PyMC version information:

I used our docker environment for testing.

Python 3.13.5 | packaged by conda-forge
>>> pm.__version__
'5.25.1+9.g09603234f'
>>> pytensor.__version__
'2.32.0'

Context for the issue:

Personally, needed to handle these warnings when writing tests. It's a bit annoying but not a show stopper.

[welcome]

]
🎉 Welcome to PyMC! 🎉 We're really excited to have your input into the project! 💖

If you haven't done so already, please make sure you check out our Contributing Guidelines and Code of Conduct.