[VOL-4388] Fix ONOS-classic statefulset podAntiAffinity configuration

[teone]

See https://jira.opencord.org/browse/VOL-4388 for a full description of the issue

[charlesmcchan]

I am wondering why we have not seen this in SD-Fabric. @hwchiu any insights?

[hwchiu]

@charlesmcchan I check one of the sd-fabric pod and we do have the same issue.

╰─$ kc -n tost get pods -o wide
NAME                                                         READY   STATUS    RESTARTS   AGE     IP              NODE     NOMINATED NODE   READINESS GATES
onos-tost-atomix-0                                           1/1     Running   0          3h37m   10.33.40.144    node1    <none>           <none>
onos-tost-atomix-1                                           1/1     Running   0          3h37m   10.33.111.238   node3    <none>           <none>
onos-tost-atomix-2                                           1/1     Running   0          3h37m   10.33.104.44    node2    <none>           <none>
onos-tost-onos-classic-0                                     1/1     Running   0          3h37m   10.33.40.139    node1    <none>           <none>
onos-tost-onos-classic-1                                     1/1     Running   0          3h37m   10.33.40.152    node1    <none>           <none>
onos-tost-onos-classic-2                                     1/1     Running   0          3h37m   10.33.40.183    node1    <none>           <none>
onos-tost-onos-classic-onos-config-loader-84d74bdbff-cnbz9   1/1     Running   0          3h37m   10.33.40.179    node1    <none>           <none>

Three instances have been deployed into the same node. I believe we haven't check it carefully.
Anyway, the patch looks good to me.