Converted warning about depth_list to a note

[adcroft]

The message that a file is being created was issued as a WARNING when we all agree it should really be a NOTE. Depth_list.nc is read if it is present to avoid recomputing a sorted list, but the absence of the file is not an error and does not warrant a warning.

Changes:

• Changed WARNING to NOTE.
• Removed MOM_mesg from imports since it wasn't being used.

[codecov]

Codecov Report

Merging #414 (14134a1) into dev/gfdl (a5129ca) will decrease coverage by 0.01%.
The diff coverage is 0.00%.

❗ Current head 14134a1 differs from pull request most recent head 4086e88. Consider uploading reports for the commit 4086e88 to get more accurate results

@@             Coverage Diff              @@
##           dev/gfdl     #414      +/-   ##
============================================
- Coverage     38.17%   38.17%   -0.01%     
============================================
  Files           269      269              
  Lines         76553    76553              
  Branches      14076    14076              
============================================
- Hits          29226    29224       -2     
- Misses        42077    42079       +2     
  Partials       5250     5250              

Impacted Files	Coverage Δ
src/diagnostics/MOM_sum_output.F90	64.84% <0.00%> (ø)

... and 1 file with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[marshallward]

I'm not even sure that this should be a NOTE. The file is always created if it doesn't exist, so I'm inclined to interpret it as normal output file. For comparison, we don't issue notes for the initial creation of ocean.stats.

[Hallberg-NOAA]

These changes seem sensible to me.

[marshallward]

After discussion with @adcroft, we agreed there is no need for this message anymore and can be removed completely.

[Hallberg-NOAA]

I disagree. In the past, I have found this message to be useful when debugging certain activities related to the diagnosis of the model potential energy. We can reduce the verbosity level of this message if you would like, but it absolutely should not be removed altogether!

[marshallward]

Is it useful because a Depth_list.nc-based computation would change the answers? Or is it for another reason?

[marshallward]

I've reverted the removal. We'll take this as it is.

[marshallward]

Gaea regression: https://gitlab.gfdl.noaa.gov/ogrp/MOM6/-/pipelines/19950 ✔️

[marshallward]

The large computational cost of constructing Depth_list.nc makes it worth notifying the user, but we discussed the possibility of skipping this notification for initial runs.

But it's not urgent, so we can come back around to this in the future, if we decide to further streamline the model output.