[FIX] Improve decomission process

[McSim85]

Often, when I run the role with state: absent, the next role run with state: present will fail with different errors like described in the bug
#429

After talking to Tailscale support, that might happen if /var/lib/tailscale/ still contains the previous machine identity (auth key, id, etc).

We have to remove this folder during decommission.
Fortunately, Ubuntu package has that task inside the tailscale.postrm, (if the user rune apt purge tailscale):

$ cat /var/lib/dpkg/info/tailscale.postrm
#!/bin/sh
set -e
if [ -d /run/systemd/system ] ; then
	systemctl --system daemon-reload >/dev/null || true
fi

if [ -x "/usr/bin/deb-systemd-helper" ]; then
    if [ "$1" = "remove" ]; then
		deb-systemd-helper mask 'tailscaled.service' >/dev/null || true
	fi

    if [ "$1" = "purge" ]; then
		deb-systemd-helper purge 'tailscaled.service' >/dev/null || true
		deb-systemd-helper unmask 'tailscaled.service' >/dev/null || true
		rm -rf /var/lib/tailscale
	fi
fi

So, this is the best way to handle tailscale state removal - purge: true, because this will always be aligned with the package maintainers config.

I am not sure about other distros, though.
Probably, we need a task like this in each distro subtree:

- name: Uninstall | Delete Tailscale Package State
  ansible.builtin.file:
    path: "/var/lib/tailscale"
    state: absent

@artis3n , please let me know how you prefer to handle this.

[artis3n]

I would like to handle this differently, let's exclude adding a variable in this PR

[McSim85]

I will create another PR.
Because current changes are actively in use :)

[McSim85]

I will create another PR.
Because current changes are actively in use :)