[snapshot-controller] Fix wait for CRDs duration

Ensure the function `ensureCustomResourceDefinitionsExist` returns when
the duration specified in `retryCRDIntervalMax` flag is reached.
This works by passing a context to the backoff so that we can
timeout with context.WithTimeout. The new (correct) default duration for
the flag is 30s to give a bit more time when controller and CRD are
created at the same time.
Also limit the amount of resources returned by the list calls to 0 to
speed up the operations because we only care that the CRD exists.

deploy/kubernetes/snapshot-controller/setup-snapshot-controller.yaml

@@ -16,10 +16,11 @@ spec:
   selector:
     matchLabels:
       app.kubernetes.io/name: snapshot-controller
-  # the snapshot controller won't be marked as ready if the v1 CRDs are unavailable
-  # in #504 the snapshot-controller will exit after around 7.5 seconds if it
-  # can't find the v1 CRDs so this value should be greater than that
-  minReadySeconds: 15
+  # The snapshot controller won't be marked as ready if the v1 CRDs are unavailable.
+  # The flag --retry-crd-interval-max is used to determine how long the controller
+  # will wait for the CRDs to become available before exiting. The default is 30 seconds
+  # so minReadySeconds should be set slightly higher than the flag value.
+  minReadySeconds: 35
   strategy:
     rollingUpdate:
       maxSurge: 0