GFDL 20241127 candidate

[jiandewang]

testing of GFDL 20241127 candidate PR (mom-ocean#1647) revealed two issues in UWM
(1) 1x1 ocean failed to retain current BL results on wcoss2, most likely due to FMA as 1x1 debug job passed regression test
(2) three non-debug llvm tests on GAEA failed to retain current BL results

I am using bisect approach trying to narrow down the issue (1) at this moment.
Marshall explained to me that GFDL haven't started any llvm test for MOM6 yet.

[jiandewang]

with many tesing, I was able to narrow down the exact commit that caysed the answer change for 1x1 setting in wcoss2 (FMA related), testing patch code from Marshall now.

[jiandewang]

Dusan explained to me that the ifx cases should be considered as experimental level as he feels that there is issue on GAEA for ifx

[jiandewang]

per discussion with @junwang-noaa , we will create new baseline for the ifx cases for this PR. In the meantime we will ask GFDL to speed up the starting of ifx testing in MOM6