Update how nitrogen deposition fluxes to surface models are handled

[fvitt]

Issue Type

Other (please describe below)

Issue Description

When CTSM is updated to always receive nitrogen deposition fluxes from the coupler, as described in ESCOMP/CTSM#1900, the ndep namelist settings in drv_flds_in should be eliminated.

Another method of determining when chemistry can provide the nitrogen deposition fluxes, such a namelist parameter in atm_in or code internal to cam-chemistry.

We should also consider moving the CAM_STREAM_NDEP_DATA_FILENAME xml variable to a cam namelist parameter in atm_in.

Will this change answers?

No

Will you be implementing this yourself?

Any CAM SE can do this