Add --error-exitcode=1 to cppcheck to fail on errors

[j-rivero]

🦟 Bug fix

Fixes #446

Summary

Cppcheck was returning exit code 0 even when errors were detected, causing CI to pass when it should fail. Adding --error-exitcode=1 ensures cppcheck returns non-zero when any errors are found.

Checklist

• Signed all commits for DCO
• Added tests
• Updated documentation (as needed)
• Updated migration guide (as needed)
• Consider updating Python bindings (if the library has them)
• codecheck passed (See contributing)
• All tests passed (See test coverage)
• While waiting for a review on your PR, please help review another open pull request to support the maintainers
• Was GenAI used to generate this PR? If so, make sure to add "Generated-by" to your commits. (See this policy for more info.)

Generated-by: Opus 4.5

Note to maintainers: Remember to use Squash-Merge and edit the commit message to match the pull request summary while retaining Signed-off-by and Generated-by messages.

[azeey]

Thanks for looking into this. I was really hopeing someone would find the time to fix this. However, before we merge/release, we should check that it doesn't break CI in downstream libraries.

[j-rivero]

Running a long one here

[j-rivero]

Thanks for looking into this. I was really hopeing someone would find the time to fix this. However, before we merge/release, we should check that it doesn't break CI in downstream libraries.

Summary of downstream:

• The colcon buildfarm jobs are not installing cppcheck, so not affected at this point.
• The Jenkins jobs are not running cppcheck only GitHub Actions according to the documentation: https://gazebosim.org/docs/latest/ci/
• I tested the change in sdformat GH actions and it is making what is suppose to do, failing the action in the presence of error https://github.com/gazebosim/sdformat/actions/runs/21640840497/job/62379667509

Looking into stable branches, the problems are being silently ignore https://github.com/gazebosim/sdformat/actions/runs/21492493963/job/61918436059

[j-rivero]

I have changed gears, we can not make all cppcheck in the first pass as mandatory, too noisy with fool warnings on standard headers. The second pass without any custom configuration seems fine to enable, making sdformat to fail with some macros that needs configuration:

nofile:0:0: information: Active checkers: There was critical errors (use --checkers-report=<filename> to see details) [checkersReport]

/home/jrivero/code/gz/codecheck_ws/src/sdformat/include/sdf/Element.hh:385:5: error: There is an unknown macro here somewhere. Configuration is required. If GZ_DEPRECATED is a macro then please configure it. [unknownMacro]
    GZ_DEPRECATED(16)
    ^
/home/jrivero/code/gz/codecheck_ws/src/sdformat/src/cmd/sdf_main.cc:137:5: error: There is an unknown macro here somewhere. Configuration is required. If SDF_PROTOCOL_VERSION is a macro then please configure it. [unknownMacro]
    SDF_PROTOCOL_VERSION ")")

I want to run this on all the libraries, still on it. In draft by now.

In my mind: we should use a modern tool that can replace the cppcheck.

[azeey]

In my mind: we should use a modern tool that can replace the cppcheck.

+1. Maybe clang-tidy?