SimPEG and numpy v2

[kkappler]

Tests failing on 3.10 and 3.11, OK on 3.8 and 3.9

File simplex_helpers.pyx:1, in init discretize._extensions.simplex_helpers()

ImportError: numpy.core.multiarray failed to import (auto-generated because you didn't call 'numpy.import_array()' after cimporting numpy; use '<void>numpy._import_array' to disable if you are certain you don't need it).

More info:

File "/usr/share/miniconda/envs/aurora-test/lib/python3.10/site-packages/discretize/unstructured_mesh.py", line 7, in <module>
    from discretize._extensions.simplex_helpers import (

A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.0.2 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need to rebuild instead e.g. with 'pybind11>=2.12'.

If you are a user of the module, the easiest solution will be to
downgrade to 'numpy<2' or try to upgrade the affected module.
We expect that some modules will need time to support NumPy 2.

[kkappler]

Seems to be this line:
from mtpy.processing import RunSummary, KernelDataset

@kujaku11 is it possible to allow aurora to import these without needing discretize and deeper simpeg stuffs? When I look at the imports in mtpy.processing I don't see and deeper reference to simpeg, but the github error calls out discretize

In a pinch I can migrate these modules to aurora for the workshop.

[kujaku11]

I guess in the next release we can make simpeg an optional install. https://setuptools.pypa.io/en/latest/userguide/dependency_management.html#optional-dependencies.