fix loss of index name with empty dfs

[dougbrn]

Fixes #409

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.27%. Comparing base (464bffa) to head (eeed100).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #410   +/-   ##
=======================================
  Coverage   97.27%   97.27%           
=======================================
  Files          19       19           
  Lines        2093     2094    +1     
=======================================
+ Hits         2036     2037    +1     
  Misses         57       57           

☔ 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.

[github-actions]

Before [464bffa]	After [ae58272]	Ratio	Benchmark (Parameter)
2.61±0.1s	2.25±0.2s	~0.86	benchmarks.ReadFewColumnsS3.time_run
136M	141M	1.04	benchmarks.CountNestedBy.peakmem_run
252M	257M	1.02	benchmarks.AssignSingleDfToNestedSeries.peakmem_run
104M	105M	1.02	benchmarks.NestedFrameAddNested.peakmem_run
109M	110M	1.02	benchmarks.NestedFrameQuery.peakmem_run
107M	109M	1.02	benchmarks.NestedFrameReduce.peakmem_run
182M	186M	1.02	benchmarks.ReadFewColumnsHTTPS.peakmem_run
1.29±0.01ms	1.29±0.01ms	1.00	benchmarks.NestedFrameReduce.time_run
262M	263M	1.00	benchmarks.ReassignHalfOfNestedSeries.peakmem_run
30.8±0.9ms	30.4±1ms	0.99	benchmarks.AssignSingleDfToNestedSeries.time_run

Click here to view all benchmarks.

[hombit]

Looks good to me, but I haven't checked versus the code in the issue

[dougbrn]

I also have not specifically tested against the LSDB case (due to some local versioning snafus), but came up with a simple case independent of LSDB. This is a good change regardless but if for some reason it doesn't fix the LSDB specific case let's re-open the issue.