Fix: Five defects in the datalayer type map and workflow comparison (B20, B139, B141, B262)

[ilayfalach]

Fixes for 4 defects from the unit-test expansion log: B20, B139, B141, B262.

The work is done and committed on the local branch fix/groupS-more, cut from tests/batch33-coverage90. Nothing is pushed yet — these branches can only land after the test-expansion stack (#1067, #1083, #1084, #1085), because every fix is verified against the xfail(strict=True) test that pins it.

What each commit says

7f6db61a fix: five defects in the datalayer type map and workflow comparison (B20, B139, B141, B262)

- B20  `typeDatatypeMap` is keyed on third-party *internal* module paths --
       "pandas.core.frame.DataFrame" and "pandas.core.series.Series". pandas 3
       re-exports both at the top level, so the real full names are
       "pandas.DataFrame" and "pandas.Series" and neither key matched: a
       DataFrame fell through to the object handler and was pickled instead of
       written as parquet, silently against the policy in CLAUDE.md. Both
       spellings are now keys, so an older pandas still resolves too.
- B139 `compareDataframeConfigurations` applied changeDotToUnderscore by
       calling .replace on every entry of ret.T.columns -- which in long
       format is the integer row index, because the frame has not been
       pivoted. Every longFormat=True comparison died with "'int' object has
       no attribute 'replace'". Only the string names are renamed now.
- B141 `findAvailableName` drew its id from the counter "simulations_<group>"
       while addWorkflowToGroup draws from "<group>". The two never agreed, so
       the "available" name it handed back was -- for any group populated the
       normal way, always -- the name of a workflow that already exists.
- B262 `DataHandler_zarr_xarray.getData` opened the archive into `df` and then
       wrote `resource.attrs = JSONToConfiguration(resource.attrs)`, where
       `resource` is the store path it was handed, not the dataset. The
       dataset's attributes were never converted and a str resource raised.

Unit layer: 5591 passed, 373 xfailed, 38 failed -- one fewer than baseline,
none new.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Verification

Full unit layer run after every commit, with the failure list diffed against a baseline captured before any edit: no regressions. The 39 baseline failures are environmental (no OpenFOAM, no current hermes, no gdal locally).