Array length mismatch in shear module

[JiHuGUO]

I installed the latest version from the main branch, but I seem to have encountered a bug. In the compute_shear method, within the if block starting at line around 593, the variables k, kx, ky, and hs are filtered using the valid flags, but l and ul are not. Since they are later used together in the computation, this may lead to a size mismatch. I'm not sure whether I misunderstood something or if this is an actual issue.

[ChengJungHsu]

I came into same issue when run attached case from grainsizeanalysis_aeolis. The issue come since line 566 compute_shear.py, which iterate calculate inner height "l" but change it from a scalar to a matrix with size of 'z0new'. Interesting thing is that my env can run parabolic_dune example (2D case) successfully.

case.zip