This PR adds a "Correctness" section for Branson that details what to expect from the radiation and material conservation printouts. It also gives golds for a simple marshak wave problem after 100 cycles. Between these two checks, changes to branson can be checked to make sure energy is not lost (e.g. a particle communication error) and that the change is numerically robust (produces the same answer after millions of accumulations on a few values).