GH-44915: [C++] Add WithinUlp testing functions

[pitrou]

Rationale for this change

When testing math-related functions, we might want to check that some results are very close to an expected value, but not necessarily exactly equal.

What changes are included in this PR?

Add functions that test whether two floating-point values are within N ulps.

("ulp" stands for "unit in the last place": https://en.wikipedia.org/wiki/Unit_in_the_last_place)

Are these changes tested?

Yes.

Are there any user-facing changes?

Potentially more useful error messages.

• GitHub Issue: [C++][Testing] Add approximate equality helpers for floating-point #44915

[pitrou]

Initially mentioned in #12681 (comment)

[pitrou]

@github-actions crossbow submit -g cpp

[pitrou]

@github-actions crossbow submit -g cpp

[pitrou]

@github-actions crossbow submit -g cpp

[github-actions]

Revision: ad2bf4f

Submitted crossbow builds: ursacomputing/crossbow @ actions-5787e09ddf

Task	Status
example-cpp-minimal-build-static
example-cpp-minimal-build-static-system-dependency
example-cpp-tutorial
test-alpine-linux-cpp
test-build-cpp-fuzz
test-conda-cpp
test-conda-cpp-valgrind
test-cuda-cpp-ubuntu-20.04-cuda-11.2.2
test-cuda-cpp-ubuntu-22.04-cuda-11.7.1
test-debian-12-cpp-amd64
test-debian-12-cpp-i386
test-fedora-39-cpp
test-ubuntu-20.04-cpp
test-ubuntu-20.04-cpp-bundled
test-ubuntu-22.04-cpp
test-ubuntu-22.04-cpp-20
test-ubuntu-22.04-cpp-emscripten
test-ubuntu-22.04-cpp-no-threading
test-ubuntu-24.04-cpp
test-ubuntu-24.04-cpp-bundled-offline
test-ubuntu-24.04-cpp-gcc-13-bundled
test-ubuntu-24.04-cpp-gcc-14
test-ubuntu-24.04-cpp-minimal-with-formats
test-ubuntu-24.04-cpp-thread-sanitizer

[zanmato1984]

LGTM

[mapleFU]

why argument name in within is n_ulp, but in assertWithin is n_ulps?

[pitrou]

Hmm... no reason I'm afraid. We can open a minor PR to fix that.

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 3 benchmarking runs that have been run so far on merge-commit 0d8c8d9.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 52 possible false positives for unstable benchmarks that are known to sometimes produce them.