Add default storage class and associated node labels (fixes #52)

[pfisterer]

This PR adds a default storage class in K8S and labels the nodes for the same availability zone. This fixes #52.

[pfisterer]

This PR also

• fixes the warning "Instead of using result|success use result is success. This feature will be removed in version 2.9." and
• updates the K8S dashboard version.

@ctrlaltdel: Sorry for pushing this, too. If you want, I can create a new PR.

[pfisterer]

Closing this PR as I've continued to improve the Playbook.

@ctrlaltdel: cf. https://github.com/infraly/k8s-on-openstack/compare/master...pfisterer:master for my current changes. These include

• Add default storage class and associated node labels
• Support for Tiller-less Helm v3
• Updated nginx ingress installed via Helm
• Add support for cluster name in kubeadm
• Drain all worker nodes before deleting them (using kubectl drain)
• Run kubeadm reset (on destroy) to cleanly destroy the cluster and associated ressource (e.g., load balancers)
• Fix for adding helm repos and installing helm charts

If you like these changes, I can prepare a PR.