Fix WCRP Geophysical Variable Detection + time bounds for sea-ice (SImon/SIday)

[rhaegar325]

Summary

Sea-ice CMORisation applied only half of its curvilinear-grid transformation:
the data variable stayed on the model (nj, ni) dimensions while the supergrid
coordinates were built on (j, i), leaving orphaned lat/lon variables that
the WCRP wcrp_cmip6:1.0 Geophysical Variable Detection check counted as
data variables. The CICE time-bounds variable was also dropped, leaving a
dangling time:bounds reference.

Problem

Expected exactly 1 geophysical variable, found 3: ['lat', 'lon', 'siconc']

Three coupled causes:

1. Dimension rename dropped. seaice_dim_rename = {ni:i, nj:j, ...}, but the
   filter if k in self.ds keeps only variables. ni/nj are pure
   dimensions (no coordinate variable), so the rename was discarded and siconc
   stayed on (time, nj, ni).
2. Orphaned model lat/lon. The mapping renames TLAT/TLON -> lat/lon; these
   2-D variables have no standard_name and were left behind on (nj, ni),
   redundant with the supergrid latitude/longitude on (j, i).
3. Stale / missing time bounds. CICE writes time:bounds = "time_bounds"
   (variable time_bounds on dim d2), but the bounds machinery expects
   time_bnds, so time_bounds was never loaded and the reference dangled.

Fix (sea_ice.py)

• Rename pure dimensions too: if k in self.ds or k in self.ds.dims, so
  ni->i / nj->j apply and siconc becomes (time, j, i), aligned with the
  supergrid.
• Drop the redundant model lat/lon in update_attributes and set the
  data variable's coordinates = "latitude longitude" (clearing the stale
  "TLON TLAT time").
• Normalise the time bounds (_normalise_time_bounds): follow time:bounds
  to the real variable, rename it to time_bnds and its non-time dim to bnds,
  and add time_bounds to the load set so it is not discarded.

Files changed

File	Change
src/access_moppy/sea_ice.py	dim-rename filter, drop model lat/lon, set coordinates, _normalise_time_bounds (+52/-2)
tests/unit/test_sea_ice.py	5 new tests (+165)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 75.0%. Comparing base (85df727) to head (95d4dd2).

Additional details and impacted files

@@           Coverage Diff           @@
##            main    #404     +/-   ##
=======================================
+ Coverage   74.9%   75.0%   +0.1%     
=======================================
  Files         28      28             
  Lines       5246    5265     +19     
  Branches     963     968      +5     
=======================================
+ Hits        3929    3950     +21     
+ Misses      1092    1091      -1     
+ Partials     225     224      -1     

Flag	Coverage Δ
unit	75.0% <100.0%> (+0.1%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 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.