enable cluster-maxunavailable for capm3-misc workload clusters

Relates to sylva-projects/sylva-core#2702 (closed)

The motivation is the following: misc pipelines are seen to fail on capm3 workload cluster because there are cases where there aren't enough nodes to allow the eviction of some pods, and that enabling cluster-maxunavailable will avoid this case.

(keeping in mind that until today, since cluster-maxunavailable isn't enabled by default we have ensure that we test both the case where it is enabled and the case where it is not -- with this MR, we still have some cases where it isn't enabled: capo pipelines, and capm3 workload clusters of non-misc pipelines)

This MR is tested via sylva-projects/sylva-core!5165 (merged), but can merge independently.