Unified rc12 testing: `diags_environment_commands` doesn't seem to be set up properly for Compy

[forsyth2]

$ zppy -c tests/integration/generated/test_complete_run_compy.cfg
$ cd /compyfs/fors729/zppy_test_complete_run_output/v2.LR.historical_0201/post/scripts
$ grep "ERROR" *status
e3sm_diags_atm_monthly_180x360_aave_environment_commands_model_vs_obs_1850-1851.status:ERROR (5)
e3sm_diags_atm_monthly_180x360_aave_environment_commands_model_vs_obs_1850-1853.status:ERROR (5)
e3sm_diags_atm_monthly_180x360_aave_environment_commands_model_vs_obs_1852-1853.status:ERROR (5)
$ cat e3sm_diags_atm_monthly_180x360_aave_environment_commands_model_vs_obs_1850-1851.o553041 
Could not find conda environment: e3sm_diags_20230816
You can list all discoverable environments with `conda info --envs`.

/tmp/slurmd/job553041/slurm_script: line 121: cdscan: command not found

From the cfg:
environment_commands = "source /qfs/people/fors729/miniconda3/etc/profile.d/conda.sh; conda activate e3sm_diags_20230816" (set by "diags_environment_commands": "source /qfs/people/fors729/miniconda3/etc/profile.d/conda.sh; conda activate e3sm_diags_20230816", in tests/integration/utils.py).

[xylar]

@forsyth2, do you have leads on this problem or do you need help?

[forsyth2]

@xylar I think I've resolved this. Looks like the conda source line just needed to be updated to use mamba. (Strangely that didn't seem to cause an issue on Chrysalis or Perlmutter though).