GH-50113: [C++][Python] Fix count for sliced union arrays

[fenfeng9]

Rationale for this change

Sliced union arrays could report incorrect results from count.

For sliced union arrays, span.GetValues<int8_t>(1) and span.GetValues<int32_t>(2) already return offset-adjusted pointers. The logical null count path then indexed those pointers with span.offset + i, effectively applying the parent offset twice.
The sparse union path also checked child nullness with i instead of span.offset + i.

What changes are included in this PR?

• Fixes logical null counting for sliced sparse and dense union arrays.
• Add C++ and Python regression tests.

Are these changes tested?

Yes.

Are there any user-facing changes?

No.

• GitHub Issue: [C++][Python] Sliced unions can report incorrect null counts in count #50113

[github-actions]

⚠️ GitHub issue #50113 has been automatically assigned in GitHub to PR creator.

[pitrou]

+1, thanks a lot for this fix @fenfeng9

[pitrou]

I've rebased from git main so as to incorporate #50108

[fenfeng9]

Some tests failed, but they look unrelated to this PR.

[fenfeng9]

And thanks for the review.

[pitrou]

Some tests failed, but they look unrelated to this PR.

Indeed they do!

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit fea7897.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 11 possible false positives for unstable benchmarks that are known to sometimes produce them.