feat(zfs): add capacity metrics and Grafana dashboard

Deploy an hourly systemd timer that writes pool and dataset capacity to
node_exporter's textfile collector, since --collector.zfs exposes only
ARC/I-O/state. Add a provisioned 'ZFS - Pools & Datasets' dashboard with
per-pool totals, per-dataset used/quota and daily growth panels.

roles/zfs/tasks/main.yml

@@ -5,3 +5,7 @@
   ansible.builtin.include_tasks: pools.yml
 - name: "Setup ZFS datasets: filesystems, snapshots, volumes"
   ansible.builtin.include_tasks: datasets.yml
+
+- name: Setup ZFS capacity metrics
+  ansible.builtin.include_tasks: metrics.yml
+  when: zfs_metrics_enabled | bool

roles/zfs/tasks/metrics.yml

@@ -0,0 +1,43 @@
+---
+# ZFS capacity metrics, exposed to Prometheus through node_exporter's textfile
+# collector (see roles/node_exporter). Runs hourly as a root system timer.
+
+- name: Deploy ZFS metrics script
+  ansible.builtin.template:
+    src: zfs-metrics.sh.j2
+    dest: "{{ zfs_metrics_script_path }}"
+    owner: root
+    group: root
+    mode: "0755"
+
+- name: Deploy ZFS metrics systemd service
+  ansible.builtin.template:
+    src: zfs-metrics.service.j2
+    dest: /etc/systemd/system/zfs-metrics.service
+    owner: root
+    group: root
+    mode: "0644"
+  notify: Reload systemd
+
+- name: Deploy ZFS metrics systemd timer
+  ansible.builtin.template:
+    src: zfs-metrics.timer.j2
+    dest: /etc/systemd/system/zfs-metrics.timer
+    owner: root
+    group: root
+    mode: "0644"
+  notify: Reload systemd
+
+- name: Flush handlers before enabling the timer
+  ansible.builtin.meta: flush_handlers
+
+- name: Enable and start ZFS metrics timer
+  ansible.builtin.systemd:
+    name: zfs-metrics.timer
+    enabled: true
+    state: started
+
+- name: Collect ZFS metrics now
+  ansible.builtin.systemd:
+    name: zfs-metrics.service
+    state: started