Add zeroing of velocity to ocean init mode

[mark-petersen]

Explicitly sets normal velocity to zero at the beginning of init mode. Each test case can then overwrite the zeros with initial velocities, but most do not. In the past the unassigned arrays wrote out zeros, but this is not true now, as described in #5230.

This code is used to create MPAS-Ocean initial conditions. It is not compiled in the E3SM code base for simulations.

Fixes #5230
[BFB]

[mark-petersen]

Note that I did not use a block loop here. Block loops are no longer supported in forward mode, and we never used multiple blocks for init mode anyway, so it seemed silly to add it. We could remove block loops from the rest of the code sometime, but for this PR, I just get the array from the first block. @philipwjones do you agree?

[xylar]

I doubt @philipwjones cares if we break init mode ;-)

[philipwjones]

I care deeply and passionately ;^). But yeah, this is fine - I've been removing them as I find them (block loops that is, not velocity)

[mark-petersen]

This PR fixes the previous problem. Here with SOMA,

compass setup -p $COMPILE_DIR -w $WORKDIR -n 249 
Setting up test cases:
  ocean/soma/32km/default

cd $WORKDIR
cd ocean/soma/32km/default/initial_state
compass run

ncdump -v normalVelocity initial_state.nc |tail -n 3
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ;
}

and the forward simulation completes successfully. I tested with both gnu debug and intel debug.

[xylar]

I did testing on Chrysalis with Intel and Intel-MPI.

I tested the 32 km default SOMA test case with this branch and with master. As expected, I found that the test case fails in forward mode without these changes but was successful with them.

I also verified that the results from the pr test suite are BFB with master. The baroclinic channel decomp test case is failing in both because of #5219

[xylar]

@cbegeman and @sbrus89, this may also fix the dam-break test case.

[sbrus89]

@xylar - unfortunately, I found that the dam break still fails with this commit. Thanks for the suggestion though.

[jonbob]

passes:

• ERS.ne11_oQU240.WCYCL1850NS.chrysalis_intel
• SMS_D_Ld3.T62_oQU120.CMPASO-IAF.chrysalis_intel
• SMS_D_Ld1.ne30pg2_EC30to60E2r2.WCYCL1850.chrysalis_intel.allactive-wcprod

merged to next

[jonbob]

merged to master