[test] Add switch to imageinout_test for enabling floating point exceptions.

[stolk]

Adds new feature: floating point exceptions, to imageinout_test binary.

Description

Running imageinout_test --enable-fpe will run the test with FPEs enabled.

Currently it is catching a FP overflow in libjxl.
Once that is fixed, it may find other FPEs.

For now, this is a Linux-only feature.

Tests

$ ./imageinout_test --enable-fpe

Checklist:

• [ x] I have read the contribution guidelines.
• [ x] I have updated the documentation, if applicable. (Check if there is no
  need to update the documentation, for example if this is a bug fix that
  doesn't change the API.)
• [ x] I have ensured that the change is tested somewhere in the testsuite
  (adding new test cases if necessary).
• [ x] If I added or modified a C++ API call, I have also amended the
  corresponding Python bindings (and if altering ImageBufAlgo functions, also
  exposed the new functionality as oiiotool options).
• [ x] My code follows the prevailing code style of this project. If I haven't
  already run clang-format before submitting, I definitely will look at the CI
  test that runs clang-format and fix anything that it highlights as being
  nonconforming.

[stolk]

Updated for clang formatting.

NOTE: if I run $ make clang-format I get a lot of modified source files. More than the files I have changed. Is the formatting in the repo out of date? Or is my clang-formatter too old or too new?

[stolk]

This found an FP overflow in libjxl and a FP domain error in OpenColorIO.

PR for OpenColorIO is here: AcademySoftwareFoundation/OpenColorIO#2067

But the CLA check failed on OpenColorIO. Not sure why, because it succeeded for PRs on OpenImageIO.

I have no CLA for libjxl, but the patch is here: libjxl/libjxl@main...stolk:libjxl:fix_fpe

With jxl and ocio fixed, $ imageinout_test --enable-fpe passes.

[stolk]

I think the CI fail is unrelated to my change?

CMake Error at src/doc/cmake_install.cmake:46 (file):
  file INSTALL cannot find
  "/__w/OpenImageIO/OpenImageIO/src/doc/CHANGES-0.x.md": No such file or
  directory.
Call Stack (most recent call first):
  cmake_install.cmake:283 (include)

[lgritz]

Yes, I already fixed. Just hit the "rerun failed tests" button on the GHA page for that run.

[lgritz]

Sorry, I take it back. You need to rebase. I'll do that for you.

[lgritz]

NOTE: if I run $ make clang-format I get a lot of modified source files. More than the files I have changed. Is the formatting in the repo out of date? Or is my clang-formatter too old or too new?

Unfortunately, clang-format seems to have little changes in every major release, which is like every 6-9 months it seems. So we have to pick one, and then we lock down for a while so that we don't end up with a big pointless reformat more frequently than every couple years. Currently, our CI is using clang-format from llvm 17.

I tend to deal with any mismatch by (a) running clang-format from my editor, reformatting only the file, and sometimes only the functions, that I know I'm touching; (b) if I must clang-format the whole codebase, I'm careful not to git add anything that had spurious reformatting but that isn't intended to be part of the PR; (c) if CI still gets slightly different formatting than on my machine (usually it matches, sometimes there are small differences), I will do minor touch-ups by hand to make it pass CI.