Make sure to actually test IPv6 connectivity to our test instance in the
IPv6 jobs.

Change-Id: I7845448804e191af356e82f8ad33c563ffd8ebd5

OVS upgrade jobs failed due to assigning an IP on the br-ex interface
(which is recycled during OVS upgrade). This change introduces a bridge
and veth pair for Neutron to use.

Change-Id: Ib3bee6e810fb8d31552d4c72c2a1ccae382c51f0

Also adds gawk for timestamping.

This helps to correlate init events with failures elsewhere.

Change-Id: I22fdb683ecf9870b2d66fedd6b40b7004317130a

This is to allow CI testing of network connectivity.

Note only "primary" node gets an address on it.
Hence, "primary" becomes our fake external router.

Depends-on: https://review.opendev.org/709361

Change-Id: I05592888796107d6de95b940c42b2bff73ac0669

Separate upgrade logic to is_upgrade job var and rename
scenarios to match.

Rename "ACTION" to "SCENARIO" (as it is a scenario).

Separate testing of dashboard (aka Horizon) and increase
its timeout to 5 minutes (CentOS 7 slow as always).

Separate initialization of core OpenStack.

Use gate setup script from ./tests/

Remove useless tox setupenv.

Do not deploy Heat when not really necessary.

Change-Id: I4fca319ccc3de7188f8b7b44c9c71321e3899467