Sanity Checks on Refactor

[jacob-moore22]

Multi-material multi-physics multi-solver setup is hard, and one of the things often missed is sanity checks on all of the combinations of things that can go wrong. We need to add more robust sanity checks after setup to verify that everything that should exists does, and that the things that do exists do not clash. Here are some known examples of checks that would make our solvers more robust. Note, due to all of the "multi's-" we support, these are generally not as trivial to implement as they seem at first glance.

1. The total volume fraction of all materials in a given element is bounded by [0,1]
2. All output variables exists before executing the solver. As an example, its possible to request level set outputs without having a level set solver requested via the input file. This would be fixed by first reading the solvers then adding checks in the output parsing.
3. (fill is as we find them)