Add dycore and/or total tendency diagnostics

[nbren12]

These are useful diagnostics to have. We currently compute using the python wrapped fv3gfs in fv3net, but it would be nice to plug-in to the fortran infrastructure as well. The microphysics emulation code doesn't require the wrapped python model, and it would be nice to avoid depending on it for diagnostics only.

[nbren12]

cc @spencerkclark

[nbren12]

I don't know how to make this change. @spencerkclark do you know of an example commit/pr which modifies the dycore diags?

[spencerkclark]

This is maybe a good example PR: #111. It does something similar for the total tendency due to physics for some tracers (implemented as dynamical core diagnostics). Implementing coarse versions of them is optional.

[brianhenn]

The PR to add the fast saturation adjustment tendency diags also uses some of the approaches that would be used to get these things, I think.