Description of changes

This PR modifies the CMEPS mediator in order to be able to run Sofar's atmosphere-wave coupled ESM and adds masking options for run-time configuration (esm_run.config) that avoid need to re-build the entire Sofar ESM for testing. It solves the these issues [1,2], which are described in more detail here.

Specific notes

• wavespotter/CMEPS main branch was synced with upstream fork (ESCOMP/CMEPS) prior to this PR, in order to highlight Sofar-specific changes. The changes in the upstream repo are mostly bug fixes, and often CESM specific additions. There are a number of improvements to atm-lnd coupling and glacier run-off options, but none of these changes should impact our coupling.
• The branch that this PR is based on (msdev/cmeps), was branched off initial and extensive modifications by @StevePny in dev/buildfix that were made to modify the public CMEPS mediator to work with Sofar's ESM.
• The main addition of msdev/cmeps is to specify srcMaskValues and dstMaskValues at run-time is built into the AdvertiseFields subroutine in the mediator (med.F90) and in the med_map_routehandles_initfrom_field subroutine (med_map_mod.F90), and requires setting coupling_mode='sofar.aw' since the changes in this PR are focused on atm-wav only. It
• The proper use of the masks should be revisited when the ocean (MOM6) is brought in.

Contributors other than yourself, if any:

@StevePny

CMEPS Issues Fixed (include github issue #):

resolves esm-run-scripts#183
resolves esm-run-scripts#191

Are changes expected to change answers? (specify if bfb, different at roundoff, more substantial)

Yes. Although coupling in Sofar ESM was running (without external forcing for WW3), the output produced was empty due to the masking issue. Fields are now produced with a standard configuration, which makes srcMaskWav=1, and sets the remaining values to ispval (default-> no masking).

Any User Interface Changes (namelist or namelist defaults changes)?

No

Testing performed

Please describe the tests along with the target model and machine(s)
If possible, please also added hashes that were used in the testing
Several cases where tested, changing the values of srcMaskAtm, dstMaskAtm, srcMaskWav and dstMaskWav, which as of this PR can be changed at runtime. Setting srcMaskAtm=1 (masking land from field exported by ATM) causes a segmentation fault. The two cases tested to work, involve setting all masks to ispaval (-987987) or changing only the srcMaskWav=1 (i.e. masking wave fields (charnock) over land), which yield similar but not identical results.