Implement Monte Carlo/Discontinuous Galerkin hybrid method

[clemekay]

From the 05/07/2025 document "MC/DC Plans Through the End of CEMeNT" :

I am currently implementing a hybrid using (Q)MC and discrete ordinates as additional feature.
So far fixed source, stationary problems appear to be running locally, though I'll need to clean up and test multigroup problems.
Time dependence should be straightforward from there.
Extending to Eigenvalue problems and improving the deterministic solvers via preconditioning and GMRES is likely next.