Switch scenario jobs to nested-virt nodes

To avoid intermittent failures and job timeouts
in scenario jobs, move these jobs to nested virt nodes.

Only moving scenario jobs as those are mostly affected
by intermittent issues(SSH Tempest failures) and job
timeouts. Also nested virt nodes are provided only by a
few nodepool providers.

Also switching non-scenario jobs to use cirros uec
image to avoid kernel panic issue.

Initial tests were done with [2] and issues/improvements
are being tracked at [3].

[1] https://bugs.launchpad.net/nova/+bug/1939108
[2] https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/819590
[3] https://etherpad.opendev.org/p/neutron-ci-improvements

Related-Bug: #1952066
Related-Bug: #1953479
Change-Id: I2f78f9de1ad1dc8c34688951c6bb2d5648d5dc3f
4 files changed