GH-41630: [Benchmarking] Fix out-of-source build in benchmarks

[assignUser]

Rationale for this change

Broken benchmarks after #41455

What changes are included in this PR?

Use /tmp/arrow as build dir.

Are these changes tested?

Are there any user-facing changes?

• GitHub Issue: [Benchmarking] Out-of-source python build breaks benchmarks #41630

[assignUser]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit c7c527c. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[github-actions]

⚠️ GitHub issue #41630 has been automatically assigned in GitHub to PR creator.

[ursabot]

Benchmark runs are scheduled for commit d5a83de. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[assignUser]

@ursabot please benchmark

[ursabot]

Benchmark runs are scheduled for commit 0446006. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

[conbench-apache-arrow]

Thanks for your patience. Conbench analyzed the 7 benchmarking runs that have been run so far on PR commit 0446006.

There were 8 benchmark results indicating a performance regression:

• Pull Request Run on test-mac-arm at 2024-05-13 03:19:01Z
  • std_copy_converting (C++) with source=cpp-micro, suite=arrow-range-benchmark
  • lazy_copy (C++) with source=cpp-micro, suite=arrow-range-benchmark
• and 6 more (see the report linked below)

The full Conbench report has more details.

[kou]

+1

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 7 benchmarking runs that have been run so far on merge-commit c555488.

There weren't enough matching historic benchmark results to make a call on whether there were regressions.

The full Conbench report has more details.