connectivity: check disrupt client deployment if enabled.

The disrupt test client Pods establish connections to the disrupt server clients.
Only after the connections are established does the readiness probe
become healthy.

Currently, we do not check if the clients have become healthy before
proceeding.

This can lead to confusing CI failures, even if you do an early sanity
check, as there may not be any restarts to report right after doing
setup.

To improve this, this commit will now cause setup to wait for the
deployment to become ready. Initial failures to connect will cause
the setup step to fail.

Signed-off-by: Tom Hadlaw <tom.hadlaw@isovalent.com>