fix: take on empty struct array returns empty array

[westonpace]

Which issue does this PR close?

• Closes StructArray take returns incorrect length with empty fields #6732.

Rationale for this change

See PR

What changes are included in this PR?

Minor change to take impl for struct arrays to account for the no-fields case

Are there any user-facing changes?

No

[tustvold]

This looks good to me, although given the never-ending source of bugs I wonder if we should be altering the constructors to force specifying the length...

[alamb]

CI failure https://github.com/apache/arrow-rs/actions/runs/13617521791/job/38062531668?pr=7224 seems related to a network issue. I restarted the check

[westonpace]

Thanks @tustvold / @alamb I technically have write permissions here but I haven't done arrow-rs merges before. Do you use a merge script (I don't see one) or just hit the "squash and merge" button?

[alamb]

Thanks @tustvold / @alamb I technically have write permissions here but I haven't done arrow-rs merges before. Do you use a merge script (I don't see one) or just hit the "squash and merge" button?

We just "hit the green button" 🚀

[westonpace]

That's too easy 😆

[westonpace]

This looks good to me, although given the never-ending source of bugs I wonder if we should be altering the constructors to force specifying the length...

I agree it is probably a good idea. It would be a breaking change. I created #7247 . There was only one test case that needed modified (I don't think empty struct arrays are a common case)