multus: Fix whereabouts tolerations for tainted nodes

Two issues were found with whereabouts toleration configuration:

1. The whereabouts-reconciler DaemonSet had `effect: NoSchedule` on its
   toleration, preventing pods from running on nodes with NoExecute or
   PreferNoSchedule taints. Removed the effect restriction to match all
   other Multus DaemonSets which use a catch-all `operator: Exists`.

2. The whereabouts-controller Deployment had no tolerations at all,
   causing the pod to not be scheduled when all worker nodes are tainted.
   Added `operator: Exists` toleration.

Bug: https://redhat.atlassian.net/browse/OCPBUGS-77772

Signed-off-by: Saurab Sonigra <ssonigra@redhat.com>

Author: ssonigra

bindata/network/multus/multus.yaml

@@ -652,7 +652,6 @@ spec:
       priorityClassName: "system-node-critical"
       tolerations:
       - operator: Exists
-        effect: NoSchedule
       containers:
       - name: whereabouts
         command: [ "/bin/sh" ]
@@ -752,6 +751,8 @@ spec:
       securityContext: {}
       serviceAccountName: multus-ancillary-tools
       terminationGracePeriodSeconds: 30
+      tolerations:
+      - operator: Exists
       volumes:
         - hostPath:
             path: {{ .CNIBinDir }}