GH-40333: [Docs] Improve env var docs for ARROW_USER_SIMD_LEVEL

[amoeba]

Rationale for this change

Conversation in https://github.com/apache/arrow/issues/40333.

What changes are included in this PR?

Just tweaks to the text in docs/source/cpp/env_vars.rst.

Are these changes tested?

I rendered them locally.

Are there any user-facing changes?

Just docs here.

• GitHub Issue: #40333

[github-actions[bot]]

:warning: GitHub issue #40333 has been automatically assigned in GitHub to PR creator.

[amoeba]

@github-actions crossbow submit preview-docs

[github-actions[bot]]

Revision: f9483a7f1d6f396148c82f3e4b8cfc87281781a1

Submitted crossbow builds: ursacomputing/crossbow @ actions-9a20dad6c8

Task	Status
preview-docs

[amoeba]

Thanks @wjones127, both changes accepted as they were good improvements.

[amoeba]

Whoops. Good eye. Done and pushed.

[zanmato1984]

+1

The improvement is very nice and clear.

[amoeba]

Thanks for the review @pitrou. I proposed two suggestions, let me know if those look good to you.

[amoeba]

Thanks for taking another look at this @pitrou. I think the new paragraph you added addresses both of your most recent comments. With that, I think this is very close if not ready to be merged.

Edit: I'm going to wait to rebase until the docs build is done.

[amoeba]

@github-actions crossbow submit preview-docs

[github-actions[bot]]

Revision: 46bcac8bec4896cec739ebb1151e82f1e54362f2

Submitted crossbow builds: ursacomputing/crossbow @ actions-25dc790af5

Task	Status
preview-docs

[amoeba]

@github-actions crossbow submit preview-docs

[github-actions[bot]]

Revision: 471126f28376bdcf2793199527fe4112ac2a74e4

Submitted crossbow builds: ursacomputing/crossbow @ actions-6fac2d8e3b

Task	Status
preview-docs

[mapleFU]

Will wait for some days, and would merge it if no negative comments

[mapleFU]

Thanks all, merged!

[conbench-apache-arrow[bot]]

After merging your PR, Conbench analyzed the 5 benchmarking runs that have been run so far on merge-commit 6a3a6ba4d1de3ebac51c74ab7845b4fdf57f0c7f.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 1 possible false positive for unstable benchmarks that are known to sometimes produce them.