Fix node drain by replacing deprecated --delete-local-data flag

[jonas-be]

Currently, when running tofu destroy, the node drain fails with the following error:

module.k3s_cluster.null_resource.servers_label["cp1|environment"] (remote-exec): Connected!
module.k3s_cluster.null_resource.servers_drain["cp1"] (remote-exec): error: unknown flag: --delete-local-data
module.k3s_cluster.null_resource.servers_drain["cp1"] (remote-exec): See 'kubectl drain --help' for usage.
module.k3s_cluster.null_resource.servers_label["cp1|environment"] (remote-exec): node/cp1 unlabeled
module.k3s_cluster.null_resource.servers_label["cp1|environment"]: Destruction complete after 1s
╷
│ Error: remote-exec provisioner error
│
│   with module.k3s_cluster.null_resource.servers_drain["cp1"],
│   on .terraform/modules/k3s_cluster/server_nodes.tf line 268, in resource "null_resource" "servers_drain":
│  268:   provisioner "remote-exec" {
│
│ error executing "/tmp/terraform_902530670.sh": Process exited with status 1

This error is caused by the --delete-local-data flag, which has been deprecated and renamed to --delete-emptydir-data.

This PR renames this flag, which will fix the node drain errors.

[xunleii]

Hi @jonas-be, thanks for your PR. I will try to make a release this weekend and to change the minimal k3s version for this module (major release because of this breaking change).

[github-actions]

This pull request has been automatically marked as stale because it has not had recent activity. If the pull request still needs attention, please leave a comment and it will be reopened.