Add bootc tests

[comps]

The last commit, adding the actual test(s), is still WIP and will change, but all commits under it should be ready for review.

[comps]

I've redesigned the bootc-image-builder test quite a bit; turns out we don't really need a Blueprint (config.toml) since we can do all the configuration in a Containerfile, incl. ssh keys for testing. And the bootc-image-builder container seems to build qcow2 just fine without config.toml, so it must be just optional.

[comps]

A note for whoever reviews this in the end:

# prepare a Container file for making a hardened image
cfile = podman.Containerfile()
cfile += util.dedent(fr'''
    FROM {src_image}
    RUN dnf -y install dnf-plugins-core
    RUN dnf -y copr enable packit/OpenSCAP-openscap-2170 centos-stream-9-x86_64
    RUN dnf -y install openscap-utils
    COPY remediation-ds.xml /root/.
    RUN oscap-bootc --profile '{profile}' /root/remediation-ds.xml
''')
cfile.add_ssh_pubkey(guest.ssh_pubkey)
cfile.write_to('Containerfile')

I wanted to de-duplicate this, so it's not copy/pasted across both bootc and anaconda tests, and I implemented a function for it,

def build_hardened_image(src_image, datastream, profile):

but it turned out to be just about as complicated as the copy/pasta, or likely even more - I had to write Containerfile into a tmpfile (since library code cannot rely on CWD being temporary), and pass multiple test-specific details such as datastream xml location (with unselected rules), profile, etc.

So it didn't make things any simpler, and it just obfuscated the logic.

So I left the copy/pasta in there, because - if anything - it's much easier to debug the test(s) by commenting sections of code out, or simply by inserting a sleep + checking the generated Containerfile manually, or modifying it, or otherwise playing with it.
Hiding it in a tmpfile / library code seemed like the worse solution.