[QE] Windows tester integration tests failing randomly

[adrianriobo]

Integration tests when executed from the windows tester on gh are showing some random failures. That is not the case when they are executed downstream or on the Linux instance on openshift-ci.

My assumption is for some reason the integration test are taking a bit more time on the provisioned Windows and the timeout for Gingko is being reach (checking the time they all fail after 1h 19m, no matter on which step they are executed).

An easy fix may be allow customize the Ginkgo Timeout.