The manual trigger is able to create a pool of OCP clusters internally, where are all clusters provisioned and handled by our automation.

This works for x86 and aarch64 architectures. But for other - Z and PPC - this does not really work, so other teams cannot use the manual trigger. That means that for every job they are executing, they have to fill all the details, which is quite exhausting.

We should add a possibility for specifying a pool of clusters, based on which the jobs will be executed. Each cluster record should have API url, username, password, architecture (that should be based on the architecture we are running the manual trigger), and most importantly OCP version. It should be possible to specify any number of clusters.