Fix single-row compound datasets with references being flattened

[bendichter]

Summary

Override shape property on AbstractZarrTableDataset to return None instead of the raw zarr array shape. This prevents hdmf's objectmapper from treating single-row compound datasets as scalars and unwrapping data[0] into a flat row.

Root cause

hdmf's ObjectMapper.__get_subspec_values unwraps builder.data when shape == (1,) and data[0] is not np.void. For BuilderZarrTableDataset, this replaces the lazy wrapper with a flat row [val1, val2, GroupBuilder], causing a column length mismatch in DynamicTable.__init__.

Fix

Return None for shape on compound table datasets so the scalar unwrap is never triggered.

Test plan

• Extended icephys tutorial reads via both Zarr conversion and zindi
• All 448 tests pass

Fixes #337

🤖 Generated with Claude Code

[codecov-commenter]

Codecov Report

❌ Patch coverage is 66.66667% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 82.04%. Comparing base (6ddd957) to head (086903f).
⚠️ Report is 2 commits behind head on unified-zarr-convention.

Files with missing lines	Patch %	Lines
src/hdmf_zarr/zarr_utils.py	66.66%	1 Missing ⚠️

Additional details and impacted files

@@                     Coverage Diff                     @@
##           unified-zarr-convention     #338      +/-   ##
===========================================================
- Coverage                    82.08%   82.04%   -0.04%     
===========================================================
  Files                            5        5              
  Lines                         1423     1426       +3     
  Branches                       298      298              
===========================================================
+ Hits                          1168     1170       +2     
- Misses                         168      169       +1     
  Partials                        87       87              

☔ 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.