Bug fixes for thread in w3_new and initializing variables

[JessicaMeixner-NOAA]

Pull Request Summary

Various bug fixes discovered while working on PR #285

Description

Bug fixes include:

• Adding routines in w3_new for threading. This seems to fix some failures for ww3_ts3 that occurred in testing 285
• Added a few missing files to .gitignore
• Initialized GA0,GAN,GD0,GDN to zero. While some compilers (and combo of switch options) automatically initialized this to zero without issue, MPI_OMPH for ww3_ts3 cases when testing 285 did not. This seems to fix the issues there, however is a bug in the develop branch not in the updates in 285.

Issue(s) addressed

• No issues were created for any of these issues. All issues were discovered while debugging the other PR.

Check list

• Is your feature branch up to date with the authoritative repository (NOAA/develop)? Yes
• Make sure you have checked the checklist for a developer submitting to develop, checklist for a developer submitting to develop and updating version number: No need for version update

Testing

• How were these changes tested? Regression tests on ncep hera.intel are in the queue now
• Are the changes covered by regression tests? yes/no. All are bug fixes
• If a new feature was added, was a new regression test added? Just bugs, no new features.
• Have regression tests been run? Running right now on NCEP hera.intel
• Which compiler / HPC you used to run the regression tests in the PR?
• Please provide the summary output of matrix.comp (matrix.Diff.out, matrixCompFull.out and matrixCompSummary.out): will be attached when finished.
  Please indicate the expected changes in the outputs (excluding the known list of non-identical tests):
  No changes are expected.

[JessicaMeixner-NOAA]

no errors in the regression tests, comparison is still running. @aliabdolali they are /scratch2/NCEPDEV/climate/Jessica.Meixner/PR_WW3/PRbugs/WW3/regtests in case you need this before I get a chance to post them myself

[ukmo-juan-castillo]

The changes seem sensible to me. I will leave Chris to run some tests on the Met Office computer.

[ukmo-ccbunney]

Thanks @JessicaMeixner-NOAA - changes look good to me. Well done on finding those uninitialised variables.
I will run the regression tests here and post the results.

[JessicaMeixner-NOAA]

Only the usual tests that are not b4b on NCEP computers were different. Log files:
matrixCompFull.out.txt
matrixDiff.out.txt
matrixCompSummary.out.txt

[ukmo-ccbunney]

Just for completeness, I can confirm that the regtest all passed OK here with the usual exceptions:

**********************************************************************
********************* non-identical cases ****************************
**********************************************************************
mww3_test_03/./work_PR3_UQ_MPI_d2                     (10 files differ)
mww3_test_03/./work_PR3_UQ_MPI_d2_c                     (10 files differ)
mww3_test_03/./work_PR3_UNO_MPI_d2                     (9 files differ)
mww3_test_03/./work_PR1_MPI_d2                     (7 files differ)
mww3_test_03/./work_PR2_UNO_MPI_d2                     (9 files differ)
mww3_test_03/./work_PR3_UNO_MPI_d2_c                     (10 files differ)
mww3_test_03/./work_PR2_UQ_MPI_d2                     (9 files differ)
ww3_tp2.18/./work_TIDE_MPI                     (1 files differ)

[aliabdolali]

It looks good to me after running the regtests and reviewing the code changes.

[aliabdolali]

See the matrix.comp
I ran the matrix on hera using intel and impi
DIFF.zip
The expected non-b4b tests are:
mww3_test_03; mww3_test_07; ww3_tp2.10;ww3_tp2.15; ww3_tp2.18