Combined slope limit in SRD

[baperry2]

In State Redistribution, the slopes used to compute the local states within a merged neighborhood (sigma in the equation below) are limited individually for each state variable being redistributed.

As a result, for example, if state variables representing internal energy and density are limited differently, the resulting q state may be thermodynamically inconsistent, even to the point of having a negative temperature or other issues. For pure hydro problems with idealized properties, this doesn't completely derail the solution but can generate some flow artifacts near the EB. The effect is much more severe for multi-component mixtures with real thermochemical properties - in PeleC, we're often forced to turn off the slopes to be able to robustly run with SRD (https://github.com/AMReX-Combustion/PeleC/pull/698).

A simple (but perhaps not trivial to implement, given the code structure) approach to fix this would be to have an option to apply the slope limiter here across all redistributed quantities, as is sometimes done for other slope computations in AMReX: https://github.com/AMReX-Codes/amrex/blob/71e9f93d3b7d23ec7b4792513a3b1ca1ce6c41c3/Src/EB/AMReX_EB_StateRedistribute.cpp#L263C71-L263C80