GH-50200: [Packaging][Debian] Drop support for bookworm

[kou]

Rationale for this change

It reached EOL on 2026-06-10.

What changes are included in this PR?

Remove Debian GNU/Linux bookwrom related code.

Are these changes tested?

Yes.

Are there any user-facing changes?

Yes.

• GitHub Issue: [Packaging][Debian] Drop support for bookworm #50200

[github-actions]

⚠️ GitHub issue #50200 has been automatically assigned in GitHub to PR creator.

[raulcd]

Can we remove it from here:

arrow/dev/release/verify-release-candidate.sh

Line 217 in 1a428c9

for target in "debian:bookworm" \

and here:

arrow/dev/release/verify-release-candidate.sh

Line 238 in 1a428c9

for target in "arm64v8/debian:bookworm" \

We might want to upgrade the image we use to upload binaries, not strictly related but a follow up maybe?

arrow/dev/release/binary/Dockerfile

Line 18 in 1a428c9

FROM debian:bookworm

[kou]

Good catch!
I've removed bookworm from the verification script.
I've also changed to trixie from bookworm for binary upload Dockerfile. I'll test it when I upload binaries in the next release.

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit f0d3f17.

There was 1 benchmark result indicating a performance regression:

• Commit Run on test-mac-arm at 2026-06-18 11:48:14Z
  • tpch (R) with engine=arrow, format=native, language=R, memory_map=False, query_id=TPCH-06, scale_factor=1

The full Conbench report has more details. It also includes information about 31 possible false positives for unstable benchmarks that are known to sometimes produce them.