Setup of magnetic simulations for 2d permanent magnet

[dokken]

@Elie

cell_labels = computational_mesh.cell_data["source_label"]
cells_ferro = cell_labels == 1  # Cells where magnetization is 1

Hc  = 1000
Mc = Constant(domain, PETSc.ScalarType(Hc))

# Defining the permeability values
mu_r.x.array[cells_ferro] = muf
mu_r.x.array[~cells_ferro] = muair

This is not right. DOLFINx reorders the cells (for data locality and to work in parallel). Search for
original_cell_indices on the forum, or look for instance at: Renumbered .msh imported data giving wrong subdomains - #4 by dokken