gridding updates

[smwoodman]

Addresses #226

• fix bug to check for 'average_method' in ds[k].attrs.keys(), rather than in ds[k].attrs.values().
• calculate profile_direction using mode, and ensure is only has time dimension.
• skip gridding variable if variable includes attribute 'average_method: none' in input timeseries.
• distance_over_ground (or DISTANCE_OVER_GROUND) is never gridded.
• if a variable does not have an average_method attribute, add 'average_method = arithmetic mean' in output netcdf.
• update tests and docs

[smwoodman]

I tried to do this in some way other than adding both 'distance_over_ground' and 'DISTANCE_OVER_GROUND' to skip_vars (eg, using utils._resolve_role), but couldn't figure out how

[jklymak]

I think if we look at the OG1 files we will see an entry for DISTANCE_OVER_GROUND

  DISTANCE_OVER_GROUND:
    processing_method:
      distance_over_ground:
        latitude: LATITUDE
        longitude: LONGITUDE
    long_name:     Distance over ground flown since mission start
    units:         km

I think this should gain a processing_role: distance_over_ground and then the "default" distance over ground should also get that processing role.

[smwoodman]

Makes sense. I added processing_role: distance_over_ground to the example slocum/seaexplorer OG1 yml files, and updated the gridding code to use utils._resolve_role for dog.

and then the "default" distance over ground should also get that processing role.

I think this means to add 'distance_over_ground' to the 'known_roles' block in _get_varnames, so I added it there (original shown below). But please redirect me as needed

def _get_varnames(deployment):
...
    known_roles = {
        'time', 'latitude', 'longitude', 'pressure', 'temperature',
        'conductivity', 'salinity', 'depth', 'profile_index',
        'profile_direction', 'oxygen_concentration',
    }

[codecov]

Codecov Report

❌ Patch coverage is 90.00000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 68.11%. Comparing base (e5b5c4a) to head (f2de423).
⚠️ Report is 6 commits behind head on main.

Files with missing lines	Patch %	Lines
pyglider/ncprocess.py	90.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #248      +/-   ##
==========================================
+ Coverage   68.02%   68.11%   +0.08%     
==========================================
  Files          15       15              
  Lines        2549     2562      +13     
==========================================
+ Hits         1734     1745      +11     
- Misses        815      817       +2     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[smwoodman]

Used method to be consistent with the for k in ds.keys(): block, even though the for td, bin_stat in profile_lookup.items(): uses bin_stat as the varname instead.

Update: I also switched from bin_stat to method in the for td, bin_stat in profile_lookup.items(): block, to be consistent throughout the function.