Stop running tempest-multinode-full

The job was added to nova with change:

  05ab017907

As I explained in the mailing list [1], we should
have sufficient test coverage in other voting jobs
that run against nova changes with the use of the
tempest-full-py3, tempest-slow-py3, nova-next, and
nova-(grenade)live-migration jobs. The tempest-full
and nova-next jobs are single-node jobs which run
compute API tests and some scenario tests but no
slow tests and no live migration. The tempest-slow
job is multinode and only runs slow tests. The
live migration jobs run live migration tests over
two nodes (and nova-live-migration also runs
evacuate tests). So tempest-multinode-full is
probably mostly redundant, a waste of time and
resources, especially since it's non-voting.

[1] http://lists.openstack.org/pipermail/openstack-discuss/2019-March/004139.html

Change-Id: Id2f748d4b119002292bda45689472a68254079ee

.zuul.yaml

@@ -269,9 +269,6 @@
         - nova-tox-functional-py35
         - tempest-full-py3:
             irrelevant-files: *dsvm-irrelevant-files
-        - tempest-multinode-full:
-            voting: false
-            irrelevant-files: *dsvm-irrelevant-files
         - tempest-slow-py3:
             irrelevant-files: *dsvm-irrelevant-files
         - grenade-py3: