Refactor the openebs operator provision test case

[nsathyaseelan]

What this PR does / why we need it:

Which issue this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close that issue when PR gets merged): fixes #

Checklist

• [ ] Does this PR have a corresponding GitHub issue?
• [ ] Have you included relevant README for the chaoslib/experiment with details?
• [ ] Have you added debug messages where necessary?
• [ ] Have you added task comments where necessary?
• [ ] Have you tested the changes for possible failure conditions?
• [ ] Have you provided the positive & negative test logs for the litmusbook execution?
• [ ] Does the litmusbook ensure idempotency of cluster state?, i.e., is cluster restored to original state?
• [ ] Have you used non-shell/command modules for Kubernetes tasks?
• [ ] Have you (jinja) templatized custom scripts that is run by the litmusbook, if any?
• [ ] Have you (jinja) templatized Kubernetes deployment manifests used by the litmusbook, if any?
• [ ] Have you reused/created util functions instead of repeating tasks in the litmusbook?
• [ ] Do the artifacts follow the appropriate directory structure?
• [ ] Have you isolated storage (eg: OpenEBS) specific implementations, checks?
• [ ] Have you isolated platform (eg: baremetal kubeadm/openshift/aws/gcloud) specific implementations, checks?
• [ ] Are the ansible facts well defined? Is the scope explicitly set for playbook & included utils?
• [ ] Have you ensured minimum/careful usage of shell utilities (awk, grep, sed, cut, xargs etc.,)?
• [ ] Can the limtusbook be executed both from within & outside a container (configurable paths, no hardcode)?
• [ ] Can you suggest the minimal resource requirements for the litmusbook execution?
• [ ] Does the litmusbook job artifact carry comments/default options/range for the ENV tunables?
• [ ] Has the litmusbooks been linted?

Special notes for your reviewer: