Minor cleanup from NRL

[matusmartini]

Description of Changes:

Collection of various small changes to bring NRL branch closer to NCAR main.

1. Use assumed-size array for sfcalb. This is simpler (and preferred?) and one does not need to pass nf_albd as an argument in setalb.
2. Correct comment in sascnvn.F. Limit overshooting not to be deeper than HALF the actual cloud.
3. Remove unused variables and one unused module namelist_soilveg_ruc in radiation_surface.f.
4. Explicitly declare variables that are used in cf_gf_deep.F90

UPDATE: Remove #ifdef MPI wrappers since MPI is a hard requirement for CCPP. Please note the code would fail anyway, if MPI variable is not defined, because the use of type(MPI_COMM) is not protected by #ifdef MPI.

Tests Conducted:

None, but used in NEPTUNE.

Dependencies:

None.

Documentation:

None.

[matusmartini]

Since this PR did not get a review yet, I added one more cleanup item to the list to remove #ifdef MPI wrappers thanks to a discussion with @climbfuji. I also updated the description of this PR.

[grantfirl]

@matusmartini Could you please pull in the latest main branch once #1171 is merged and we'll get this reviewed and tested?

[matusmartini]

@matusmartini Could you please pull in the latest main branch once #1171 is merged and we'll get this reviewed and tested?

Thank you @grantfirl. I resolved the conflicts which was straightforward and merged main.

[grantfirl]

@matusmartini Please merge in matusmartini#3. We needed one more update to main and a couple of compilation error fixes.

EDIT: nevermind, I had push privileges to your branch.

[grantfirl]

This passes SCM RTs in NCAR/ccpp-scm#638 (no baseline changes).

[grantfirl]

All UFS RTs passed with this code on top of other changes in main that haven't made their way back to ufs/dev yet EXCEPT hafs_regional_atm_thompson_gfdlsf_intel. It's odd that there is a baseline change at all, although it may be a fluke (i.e. corrupted baseline). Regardless, if a failure is repeated during a PR syncing the NCAR/main fork back to ufs/dev, we can investigate at that time. I think that it is fine to merge into main ATM.

[matusmartini]

Thank you @grantfirl for fixing the issue and testing.