GH-46642: [Format] Add footnote clarifying REE layout has O(log n) random access

[amoeba]

Rationale for this change

Resolves #46642.

What changes are included in this PR?

• Updated columnar format doc

Are these changes tested?

Yes, rendered locally.

Are there any user-facing changes?

No.

• GitHub Issue: [Format] Clarify that the REE layout has O(log n) random access #46642

[amoeba]

For reviewers: I tried to keep it short though I think there are other ways we might address this such as changing the bullet to something like "Generally O(1) (constant-time) random access". Feel free to suggest a different direction.

[kou]

@github-actions crossbow submit preview-docs

[github-actions]

Revision: ed91ace

Submitted crossbow builds: ursacomputing/crossbow @ actions-39b42b1501

Task	Status
preview-docs

[kou]

+1

[amoeba]

Merged. Thanks for the quick reviews all.

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit 53b4af4.

There were 65 benchmark results with an error:

• Commit Run on amd64-m5-4xlarge-linux at 2025-06-03 01:18:40Z
  • tpch (R) with engine=arrow, format=parquet, language=R, memory_map=False, query_id=TPCH-04, scale_factor=10
  • tpch (R) with engine=arrow, format=native, language=R, memory_map=False, query_id=TPCH-12, scale_factor=1
• and 63 more (see the report linked below)

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 13 possible false positives for unstable benchmarks that are known to sometimes produce them.