Add example that illustrates how to use submeshes with DOLFINx-adjoint

[jorgensd]

Summary

• Solves the EMI equations, as written out in https://github.com/scientificcomputing/fenics-in-the-wild/blob/main/src/ucs/emi/emi_primal_single.py, with the discretization by MiroK (see added reference).
• Uses a Poisson-mother type of solution

Bug-fixes/minor improvements

• Add naming to constant function constuctor (as it becomes a function).
• Minor fix for none-blocks in pad blocks by parts
• In assemble_scalar only add dependencies for OverloadedType. Note added in current demo regarding this behavior.

The demo is generated with the help of Claude, using the aforementioned sources as input (the fenics-in-the-wild, dolfinx-adjoint, dolfinx-adjoint-knowledge and scifem repos).
The bug-fixes is my own work, no AI involved.

Things to consider:

In the demo it states:

# This BC value is deliberately a plain `dolfinx.fem.Constant`, not a
# `dolfinx_adjoint.Constant`, unlike the physical parameters above: tracking it (via
# `dolfinx_adjoint.dirichletbc`) breaks the adjoint gradient for this particular
# entity_maps + blocked LinearProblem combination -- confirmed with a Taylor test, whose
# rate drops from the correct ~1.0 to ~-1.4 as soon as the BC value is annotated. Since
# the BC is fixed data, not a control, leaving it untracked is also the right modelling
# choice, not just a workaround; the discrepancy is worth a closer look/report upstream.

which hopefully should be resolved with #83.