flexible restart writing in UFS

[jiandewang]

This mini PR add the capability of flexible restart write times which is controlled by "restart_fh" for all subcomponents (ocean, ice, wave and mediator) in UFS. Detail can be found at NOAA-EMC#139
The only file that is being modified is mom_cap.F90
Thanks for @alperaltuntas to pre-test this in NCAR system

[abozec]

COAPS is not using the nuopc_cap so we approve ...

[Hallberg-NOAA]

In general, we are not very happy with the use of CPP macros, as it leads to code that might not even compile, but in this case this is an extension of macros that already exist and are in use to accommodate differences between the CESM version of the NUOPC coupler and the version used by other people. We might want to think more carefully about whether there are other non-EMC and non-CESM folks who are using the NUOPC coupler and might need to be consulted with these changes.

That being said, GFDL does not use the NUOPC cap for our configurations, so this would not be a problem from our end. (@marshallward will need to issue the formal approval from the GFDL side.)

[marshallward]

@Hallberg-NOAA Your approval counts towards the total approval, so nothing more is needed from GFDL.

Although it's "one vote per node", there's no way to enforce this in GitHub, so centers with multiple approvers have to self-enforce this. (I have removed myself from the reviewer list.)