Add tidal boundary conditions to split-explicit time integrator

[cbegeman]

This PR adds the option to have a tidal boundary condition with the split-explicit time integrator. Formerly, only tidal boundary conditions were supported with RK4. The changes to the ssh and layerThickness terms are described in https://www.overleaf.com/read/htysppyxksyh#6b2d8f.

This PR also includes options for a linearly-varying tidal boundary condition to match a test case from O'Dea et al. 2021 used for validation.

This work depends on #6047 in that the test cases used for verification of the forcing will fail without the use of upwinded advection and that fix.

[BFB] when feature is turned off

This feature is only available in MPAS-Ocean standalone configurations.

[cbegeman]

Testing

This PR is BFB with #6047 across the compass PR test suite on chrysalis with intel, openmpi.

The linear tidal forcing feature is tested with a new configuration of the drying_slope test case following O'Dea et al. (2021) on chrysalis with intel, openmpi in which the tide drops linearly by 1m over 3 hours.

[rljacob]

notes: ready but low priority

[jonbob]

@cbegeman -- I asked @mark-petersen to review, but please add anyone else who makes sense as well

[mark-petersen]

@cbegeman since #6047 is now merged, I rebased locally on master (09491ec) and I'm testing that. The rebase had no conflicts.

[mark-petersen]

Compiled with intel on chrysalis and ran nightly test suite. Compares bfb against master branch point, as expected.

This all looks reasonable. The new code is protected by flags when the boundary tidal forcing is off. Approving based on @cbegeman's tests above, where the drying slope test has boundary forcing. Very nice!

[jonbob]

passes:

• SMS_Ld1.ne30pg2_EC30to60E2r2.WCYCL1850.chrysalis_intel.allactive-wcprod_1850
• ERS_Ld5.T62_oQU120.CMPASO-NYF.chrysalis_intel

merged to next

[cbegeman]

@mark-petersen Thank you for your review and testing and @jonbob thanks for your testing!

[jonbob]

merged to master