Investigate Flakey Tests in E2E CI

[aiyengar2]

The following dashboard queries always seem to be flakey in E2E CI:

db/kubernetes-compute-resources-namespace-pods/IOPS(Reads+Writes)_query0
db/kubernetes-compute-resources-namespace-pods/Current_Storage_IO_query0
db/kubernetes-compute-resources-namespace-pods/Current_Storage_IO_query1
db/kubernetes-compute-resources-namespace-pods/Current_Storage_IO_query2

db/kubernetes-compute-resources-pod/IOPS_query0
db/kubernetes-compute-resources-pod/IOPS_query1
db/kubernetes-compute-resources-pod/IOPS(Reads+Writes)_query0
db/kubernetes-compute-resources-pod/Current_Storage_IO_query0
db/kubernetes-compute-resources-pod/Current_Storage_IO_query1
db/kubernetes-compute-resources-pod/Current_Storage_IO_query2

db/kubernetes-compute-resources-project/IOPS(Reads+Writes)_query0
db/kubernetes-compute-resources-project/Current_Storage_IO_query0
db/kubernetes-compute-resources-project/Current_Storage_IO_query1
db/kubernetes-compute-resources-project/Current_Storage_IO_query2

In #39, these tests are being automatically skipped but we should ideally investigate why these dashboards tend to have no data in random runs and put in fixes for them.

[aiyengar2]

In addition to fixing these flakey tests, we should also introduce a step into CI to perform a helm install of Longhorn before creating the ProjectHelmChart so that we can trigger the ProjectHelmChart to enable persistent storage for Prometheus and Grafana; this will ensure that we can validate that the Persistent Volume Metrics that are currently being skipped can be validated

[aiyengar2]

We should also investigate the issues in https://github.com/rancher/prometheus-federator/actions/runs/3734091815/jobs/6335719834; this is probably related to rancher/rancher#39430