Create a summary type that is outside of both soil-BGC and BGC that holds the total variables

In #1959 we needed to move the summary variables (total C and N) from biogeochemistry to soilbiogeochemistry, so it could be used by both FATES and non-FATES. A further refinement would be to have a type that just contains the summary type operation and the data needed for it. This is a design question to make the code easier to read and understand and put related data together.

The types as they are distinguish between Carbon and Nitrogen, but this Summary type would maybe include both total-C and total-N in the same type? Right now there is a Summary method for CarbonState and CarbonFlux as well as NitrogenState and NitrogenFlux. It seems like all four should NOT be combined together, but maybe some of them would?

Variables summed in CarbonState%Summary include: total-C, total-ecosystem-C, litter, SOM, and decomp-Cpools.

This is related to #1992. This type should probably reside in the directory created in that case that's a bridge between these two directories.