Those annotations create enormous clutter in diff views on GitHub, making PR reviews complicated.
Since we are currently not headed toward 100% coverage, they provide little benefit.
The macos-latest image recently upgraded to macOS 12. This includes
the iOS 16 SDK which deprecates some features which we're using. The
short term fix is to just keep using macOS 11 for 2.6.x.
In order to prevent an unnecessary modeswitch and use the current TTY resolution, don't use the preferred mode reported by DRM. This is useful for people who set a custom resolution at boot and are forced into the monitor's preferred resolution when a SFML DRM app starts
This ensure that tools like run-clang-tidy don't try to analyze them
and better mimmicks how Doctest would behave if we were depending on
Doctest via find_package instead.
Skip Android, iOS, and Framework builds simply because I can't get
them to work and don't want that holding up getting the rest of the
install tests merged.
Skip the Static DRM install test because there's a bug in the install
that needs to be fixed.
libc++ is already the default and GCC can't even be used so there's
no circumstance where we'd need to explicitly tell Clang to use
libc++. I confirmed that even with this removed, libc++ headers are
still being used and found.
You cannot buy anything but a 64 bit x86 or ARM Mac today so there
is no need to ensure that one of these two architectures is being
used. It's definitely one of the two.
Test output now shows the module name left-aligned so you can easily
browse to see what modules have what tests and know what module a
failing test is coming from.
Using a newer CMake version allows us to make use of the
MSVC_RUNTIME_LIBRARY added in CMake 3.15, which handles static linking
of the runtime library with MSVC a lot simpler