[Release] Fixed file path for FIXgsm

[EdwardSnyder-NOAA]

DESCRIPTION OF CHANGES:

Added back the fix_dir_input_grid variable to the exregional_make_ics.sh and exregional_make_lbcs.sh scripts.

TESTS CONDUCTED:

Ran test on Hera and more tests will be conducted through the Jenkins pipeline.

DEPENDENCIES:

None.

DOCUMENTATION:

No documentation changes needed.

ISSUE (optional):

CONTRIBUTORS (optional):

@mark-a-potts

[mark-a-potts]

This looks good. I am not sure we need the parameter in exregional_make_ics.sh, but we can test it to make sure it doesn't break anything. Can you also switch over all the "develop" paths in ush/machine/*.sh to "v2p0" and we can combine two fixes into one?

[EdwardSnyder-NOAA]

This looks good. I am not sure we need the parameter in exregional_make_ics.sh, but we can test it to make sure it doesn't break anything. Can you also switch over all the "develop" paths in ush/machine/*.sh to "v2p0" and we can combine two fixes into one?

Made the changes as requested.

The review has been addressed

[gsketefian]

@EdwardSnyder-NOAA @mark-a-potts Just curious if you know whether this line was necessary to fix the make_lbcs failures. If it was, it means something must have changed either in chgres_cube or the fix files. (I always thought the external model data files already contained a description of the external model's grid.) To keep the namelist as simple as possible, it would be good to remove it if it's not really needed. Not necessarily for this PR; I'm thinking more about when we move these changes into develop. Thanks.

[mark-a-potts]

From our testing, it was definitely needed. I am not sure that it is needed for the exregional_make_ics.sh script, though. Without that line in the namelist, chgres_cube was looking for NULL/latlon.grid3.nc (or something very similar) and segfaulting when it could not open the file.

[gsketefian]

After digging into this a bit more, looks like this line was removed very recently from both files in PR #775. I guess that was needed due to a change in UFS_UTILS (not sure what), but the release/public-v2 branch still needs the lines (although I haven't actually tested the develop branch with the two tests that were failing in the release branch to see if it works; maybe I'll give that a shot now).