BPF CI regression in IMA

[David Vernet <void@xxxxxxxxxxxxx>]

Hi Matt,

It seems that [0] has caused BPF CI runs to start failing in the IMA
tests. See [1] for an example of a failed run.

[0]: https://lore.kernel.org/all/Y7T1hEhIL5TEmLEN@xxxxxxxxxx/
[1]: https://github.com/kernel-patches/bpf/actions/runs/4356022302/jobs/7614059918#step:6:6298

CC'ing KP and Robert who added the test.

Thanks,
David