Longevity test cronjobs fail to re-rollout backend eployments

[pleshakov]

The longevity test deploys a cronjobs to re-rollout backend deployments
https://github.com/nginxinc/nginx-gateway-fabric/blob/770ff04e3f1a77e5cd323a5fcb52ec3fbae882da/tests/suite/longevity_test.go#L28
https://github.com/nginxinc/nginx-gateway-fabric/blob/770ff04e3f1a77e5cd323a5fcb52ec3fbae882da/tests/suite/manifests/longevity/cronjob.yaml

However, they fail to re-rollout.

Example log:

2024-03-14 08:59:00.823 EDT
{
2024-03-14 08:59:00.823 EDT
  "kind": "Status",
2024-03-14 08:59:00.823 EDT
  "apiVersion": "v1",
2024-03-14 08:59:00.823 EDT
  "metadata": {},
2024-03-14 08:59:00.823 EDT
  "status": "Failure",
2024-03-14 08:59:00.823 EDT
  "message": "deployments.apps \"coffee\" is forbidden: User \"system:serviceaccount:longevity:rollout-mgr\" cannot patch resource \"deployments\" in API group \"apps\" in the namespace \"default\"",
2024-03-14 08:59:00.823 EDT
  "reason": "Forbidden",
2024-03-14 08:59:00.823 EDT
  "details": {
2024-03-14 08:59:00.823 EDT
    "name": "coffee",
2024-03-14 08:59:00.823 EDT
    "group": "apps",
2024-03-14 08:59:00.823 EDT
    "kind": "deployments"
2024-03-14 08:59:00.823 EDT
  },
2024-03-14 08:59:00.823 EDT
  "code": 403
2024-03-14 08:59:00.823 EDT
* Connection #0 to host kubernetes.default left intact
2024-03-14 08:59:00.826 EDT
}