Add 'ready' entrypoint for exec-based readiness checks

[andrewstucki]

This is currently stacked on the PR wiring up the ready probe for the sidecar, but it introduces an entrypoint for also issuing a single exec-based readiness check using the same underlying prober (but with the more relaxed IsClusterBrokerReady rather than executing IsClusterBrokerHealthy). Adding this as after talking with @birdayz we discussed that for the current v1-based nodes in cloud the constrained resources would likely not work well with provisioning an additional sidecar that would sap resources from the main container.

Instead they can exec the probe directly in the main container and we move ever so closer to having more shared components between the v1/v2 implementation.

Once #379 lands I can adjust the base branches here.

[RafalKorepta]

Adding this as after talking with @birdayz we discussed that for the current v1-based nodes in cloud the constrained resources would likely not work well with provisioning an additional sidecar that would sap resources from the main container.

I almost agree with this statement. The exec probe would eat resources too regardless of it's place. The issue of specifying another container that should have request and limits might be problematic in cloud environment.

[andrewstucki]

@RafalKorepta sorry, yes, I should have worded that more carefully, the resource limits is the main concern.