MPAS-SI crashes on small timesteps

[mkstratos]

See issue #5072: "MPAS-SI crashes when running in thermodynamic mode (F cases) with small time steps." When dtime=1, as used in the time step convergence (TSC) test, an issue occurs with MPAS-SI at a particular time step.

It seems to depend on input data, as for different initial atmosphere / land conditions causes the problem to appear at different time steps (https://github.com/E3SM-Project/E3SM/issues/5072#issuecomment-1191829234).