"use_cn" is used inconsistently in soilbiogechem for FATES than in the rest of the model

For most of the model the logical namelist variable "use_cn" is used as a flag to say that the above ground Carbon-Nitrogen model is turned on. However, when FATES is on use_cn=.false. is used in the soil biogeochemistry to signal that Nitrogen isn't in use.

Here's a grep of the code for soil-BGC:

``` fortran
SoilBiogeochemLittVertTranspMod.F90: use clm_varctl , only : iulog, use_c13, use_c14, spinup_state, use_vertsoilc, use_fates, use_cn
SoilBiogeochemLittVertTranspMod.F90: if (use_cn ) then
SoilBiogeochemPrecisionControlMod.F90: use clm_varctl , only : iulog, use_c13, use_c14, use_nitrif_denitrif, use_cn
SoilBiogeochemPrecisionControlMod.F90: if (use_cn) then
SoilBiogeochemPrecisionControlMod.F90: if (use_cn) then
SoilBiogeochemStateType.F90: use clm_varctl , only : use_vertsoilc, use_cn
SoilBiogeochemStateType.F90: if (use_cn) then
```

The thin that's done in LittVertTransfer is...

``` fortran
!------ loop over litter/som types
do i_type = 1, ntype

select case (i_type)
case (1) ! C
conc_ptr => soilbiogeochem_carbonstate_inst%decomp_cpools_vr_col
source => soilbiogeochem_carbonflux_inst%decomp_cpools_sourcesink_col
trcr_tendency_ptr => soilbiogeochem_carbonflux_inst%decomp_cpools_transport_tendency_col
case (2) ! N
if (use_cn ) then
conc_ptr => soilbiogeochem_nitrogenstate_inst%decomp_npools_vr_col
source => soilbiogeochem_nitrogenflux_inst%decomp_npools_sourcesink_col
trcr_tendency_ptr => soilbiogeochem_nitrogenflux_inst%decomp_npools_transport_tendency_col
endif
```

So it's basically used to tell soil-BGC that FATES isn't using Nitrogen.

This is first of all confusing to use it for two purposes like this. Second, FATES is going to be adding nutrients in via PARTEH and one of the first nutrients will be Nitrogen. So there should be a different flag that can be used to tell soil-BGC that Nitrogen is on or off.

In above ground BGC the supplemental Nitrogen flag supl_nitro is used to say that Nitrogen is unlimited. But, that flag is currently incompatible with the new CLM5.0 Nitrogen options, so it's use should be limited.

As such, I think soil-BGC should have a different flag that's set internal to the model to determine if soil-BGC is using Nitrogen or no. Maybe something like soilbgc_nitro? It probably shouldn't be a namelist item, because it isn't something that a user would toggle on or off.

@rgknox @glemieux @bishtgautam