fails to build in rawhide due to libdnf cmake compat

[cgwalters]

see https://download.copr.fedorainfracloud.org/results/@CoreOS/continuous/fedora-rawhide-x86_64/08706783-rpm-ostree/builder-live.log.gz

  CMake Error at CMakeLists.txt:7 (cmake_minimum_required):
    Compatibility with CMake < 3.5 has been removed from CMake.

    Update the VERSION argument <min> value.  Or, use the <min>...<max> syntax
    to tell CMake that the project requires at least <min> but has been updated
    to work with policies introduced by <max> or earlier.

    Or, add -DCMAKE_POLICY_VERSION_MINIMUM=3.5 to try configuring anyway.



  thread 'main' panicked at /builddir/build/BUILD/rpm-ostree-2025.5.22.gc86c180e-build/rpm-ostree-2025.5.22.gc86c180e/vendor/cmake/src/lib.rs:1115:5:

  command did not execute successfully, got: exit status: 1

  build script failed, must exit now
  note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
warning: build failed, waiting for other jobs to finish...
make[2]: *** [Makefile:4729: cargo-build] Error 101
make[2]: Leaving directory '/builddir/build/BUILD/rpm-ostree-2025.5.22.gc86c180e-build/rpm-ostree-2025.5.22.gc86c180e'

[jlebon]

Looks like https://github.com/rpm-software-management/libdnf/issues/1699 is fixed now. Doing a naive submodule bump definitely shows a bunch of APIs that will need to be adjusted. Also unsure about the risk. Tempting to just have a libdnf branch (I guess on a fork somewhere?) where we cherry-pick just what we need.

[cgwalters]

Yeah last time this came up I think the core problem was this https://github.com/coreos/rpm-ostree/pull/5140#issuecomment-2452588932

[cgwalters]

And I guess now, we could just eat the cost of having a rhel9 fork. It probably wouldn't be too bad, and there isn't that much happening here nowadays that the cost of cherry picking would be too high.

(And I guess if we do a rhel9 fork, why not rhel10 too? Then it unblocks being more aggressive with libdnf sync for fedora)