Support nvfortran compiler

[kvrigor]

nvfortran comes from the NVIDIA HPC SDK.

[kvrigor]

nvfortran succeeds in building eCLM standalone. However there's a compile error when building in coupled mode:

[ 87%] Building Fortran object clm5/CMakeFiles/clm.dir/oasis3/oas_sendReceiveMod.F90.o
nvfortran-Fatal-/p/software/fs/jurecadc/stages/2025/software/NVHPC/25.5-CUDA-12/Linux_x86_64/25.5/compilers/bin/tools/fort1 TERMINATED by signal 11
gmake[5]: *** [clm5/CMakeFiles/clm.dir/build.make:220: clm5/CMakeFiles/clm.dir/oasis3/oas_sendReceiveMod.F90.o] Error 2

I need to investigate further what's causing this odd error.

[kvrigor]

The error above seems to be already fixed for some reason.. @AGonzalezNicolas could you also help to verify?

ENV_FILE="env/jsc.2025.nvhpc.openmpi"
# - or -
# ENV_FILE="env/jsc.2025.nvhpc.psmpi"

cd TSMP2
git pull && git checkout env-nvhpc

./build_tsmp2.sh eclm --clean_first --env $ENV_FILE
./build_tsmp2.sh eclm parflowgpu --clean_first --env $ENV_FILE

@jjokella you may also be interested in testing eCLM-PDAF builds with this new compiler

[AGonzalezNicolas]

@kvrigor env jsc.2025.nvhpc.openmpi is build correctly for eclm with no error message.
On the other hand, combination eclm-parflowGPU gives this error:

[ 87%] Building Fortran object clm5/CMakeFiles/clm.dir/oasis3/oas_defineMod.F90.o
[ 87%] Building Fortran object clm5/CMakeFiles/clm.dir/oasis3/oas_sendReceiveMod.F90.o
nvfortran-Fatal-/p/software/fs/jurecadc/stages/2025/software/NVHPC/25.5-CUDA-12/Linux_x86_64/25.5/compilers/bin/tools/fort1 TERMINATED by signal 11
gmake[5]: *** [clm5/CMakeFiles/clm.dir/build.make:220: clm5/CMakeFiles/clm.dir/oasis3/oas_sendReceiveMod.F90.o] Error 2
gmake[4]: *** [CMakeFiles/Makefile2:342: clm5/CMakeFiles/clm.dir/all] Error 2
gmake[3]: *** [Makefile:136: all] Error 2
gmake[2]: *** [CMakeFiles/eCLM.dir/build.make:112: eCLM/src/eCLM-stamp/eCLM-install] Error 2
gmake[1]: *** [CMakeFiles/Makefile2:113: CMakeFiles/eCLM.dir/all] Error 2
gmake: *** [Makefile:91: all] Error 2

[jjokella]

For eCLM-PDAF, see HPSCTerrSys/TSMP2#116

[kvrigor]

Previous error is gone when using nvfortran in Stages/2026. @AGonzalezNicolas @jjokella this PR is ready for review :)

JURECADC_NVHPC-PSMPI_eCLM_2026-01-28_08-54.log
JURECADC_NVHPC-OPENMPI_eCLM_2026-01-16_11-19.log