Replace collect-profile jobs that haven't completed

Problem: If for some reason the collect-profiles CronJob is unable to
finish for an extended period of time, the number of pods created
when the jobs are able to run can exceed node pod quota limits.

Solution: Set the collect-profiles cronJob's spec.concurrencyPolicy to
"Replace" so that only one active collect-profiles pod  exists at any
time.

Author: awgreene

manifests/0000_50_olm_07-collect-profiles.cronjob.yaml

@@ -8,6 +8,7 @@ metadata:
   namespace: openshift-operator-lifecycle-manager
 spec:
   schedule: "*/15 * * * *"
+  concurrencyPolicy: "Replace"
   jobTemplate:
     spec:
       template:

scripts/generate_crds_manifests.sh

@@ -247,6 +247,7 @@ metadata:
   name: collect-profiles
   namespace: openshift-operator-lifecycle-manager
 spec:
+  concurrencyPolicy: "Replace"
   schedule: "*/15 * * * *"
   jobTemplate:
     spec: