This adds a rough idea of what the pagure commit consumer could look
like.
We are once again misusing the item to carry additional information.
I'd really like to be able to have libtaskotron take the arbitrary input
(and I think we will have to do this sooner than later), but it is out
of the scope for this diff
This also fixes a heisenbug in testing, where depending on the order in
which the tests ran, the dist_git_commit/pagure_git_received were getting
either [x86_64] or [i386, x86_64] as valid arches, because some of the
testsuites change the default config value and do not clean up.
I'm being picky, but can you fix this? I don't see a reason why that has to be an "end of line" commit