Unsupported array dtype in compute forward w/ HCP data

Describe the bug

When computing a forward solution on HCP MEG data (following the tutorial), Numba complains about a dtype.

Steps to reproduce

I'm copying and pasting code from the MNE-HCP tutorial here

https://mne.tools/mne-hcp/auto_tutorials/plot_compute_forward.html#sphx-glr-auto-tutorials-plot-compute-forward-py

Expected results

I expected the mne.make_forward_solution call to succeed.

Actual results

>>> fwd = mne.make_forward_solution(info, trans=head_mri_t, bem=bem_sol, src=src_subject)
...
TypingError: Failed in nopython mode pipeline (step: nopython frontend)
non-precise type pyobject
During: typing of argument at /Users/duke/miniforge3/lib/python3.9/site-packages/mne/forward/_compute_forward.py (132)

File "../../miniforge3/lib/python3.9/site-packages/mne/forward/_compute_forward.py", line 132:
def _do_lin_field_coeff(bem_rr, tris, tn, ta, rmags, cosmags, ws, bins):
    <source elided>
    """
    coeff = np.zeros((bins[-1] + 1, len(bem_rr)))
    ^ 

This error may have been caused by the following argument(s):
- argument 1: Unsupported array dtype: >i4

I guess this may be an m1 arm64 issue about integer sizes but not sure. This is Numba running natively and not through Rosetta 2 x86_64 emulation.

Additional information

Platform:         macOS-12.3.1-arm64-arm-64bit
Python:           3.9.7 | packaged by conda-forge | (default, Sep 29 2021, 19:24:02)  [Clang 11.1.0 ]
Executable:       /Users/duke/miniforge3/bin/python3
CPU:              arm: 10 cores
Memory:           Unavailable (requires "psutil" package)
mne:              1.0.2
numpy:            1.20.3 {blas=NO_ATLAS_INFO, lapack=lapack}
scipy:            1.7.3
matplotlib:       3.5.0 {backend=module://matplotlib_inline.backend_inline}

sklearn:          1.0.1
numba:            0.54.1
nibabel:          3.2.2
nilearn:          Not found
dipy:             Not found
cupy:             Not found
pandas:           1.3.4
pyvista:          Not found
pyvistaqt:        Not found
ipyvtklink:       Not found
vtk:              Not found
PyQt5:            Not found
ipympl:           Not found
pooch:            v1.6.0

mne_bids:         Not found
mne_nirs:         Not found
mne_features:     Not found
mne_qt_browser:   Not found
mne_connectivity: Not found

Start by running the HCP compute-forward tutorial and reproducing the failure at mne.make_forward_solution. Inspect mne/forward/_compute_forward.py, especially _do_lin_field_coeff and the dtype passed to it. Done means the tutorial's forward-solution call succeeds on the reported arm64 setup without the Numba dtype error.