Materials burnt with zero fluxes in openmc.deplete leading to improperly finalized depletion_results.h5

[yrrepy]

Description

Working with openmc.deplete.get_microxs_and_flux
if it is using a mesh tally, if the results has mesh tally bins with energy-integrated results of zero (deep shielding without sufficient VR), the statepoint with fluxes and the microXS are prepared and finalized.

This is then packed off to the operator and the solver.
A pile of warnings are received in the integrate solve, but it finishes successfully.

py312/lib/python3.12/site-packages/openmc/deplete/cram.py:82: MatrixRankWarning: Matrix is exactly singular
  y += 2*np.real(alpha*sla.spsolve(A - theta*ident, y))
py312/lib/python3.12/site-packages/uncertainties/core.py:1024: UserWarning: Using UFloat objects with std_dev==0 may give unexpected results.
  warn("Using UFloat objects with std_dev==0 may give unexpected results.")

However, the resulting depletion_results.h5 is not correctly finalized and is not usable.

Enhancement:
Allowing fluxes of zero in the openmc.deplete solver (or at least for proper finalization of the depletion_results.h5 when this is the case).

Alternatives

Coding to redact from the material list, flux list and microxs lists those material zones with tallies that are zero from being passed to the operator.