feat(host_agent): each drive's health, read with smartctl, alertable (#5789)

The root helper gains one read-only action, `drives`: it finds the drives
itself (smartctl --scan-open), reads each as JSON, and writes a summary the
agent carries on every sample (agent 1.20.0). Steward judges the figures in
drives.py (failing / worn / watch / ok / no SMART / unreadable), records them
per drive and as host roll-ups for covering rules, keeps the newest report on
the registration and lists it on the host page.

install-deps.sh installs smartmontools with apt, dnf, yum, zypper, pacman or
apk, run by install.sh and every agent playbook; one it can't install is
reported, never fatal.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

plugins/host_agent/templates/_host_metrics.html

@@ -119,6 +119,41 @@
 </div>
 {% endif %}
 
+{# ── Drive health (#5789): each drive as the root helper read it with
+   smartctl, hourly; what needs attention first (drives.py judges them). ──── #}
+{% set drive_badge = {"failing": ("crit", "Failing"), "unreadable": ("warn", "Unreadable"),
+                      "worn": ("warn", "Worn"), "watch": ("warn", "Watch"), "ok": ("ok", "OK"),
+                      "unavailable": ("inactive", "No SMART")} %}
+{% if drives or drives_error %}
+<div class="card">
+  <div class="section-title" style="margin-bottom:0.5rem;display:flex;justify-content:space-between;gap:0.75rem;">
+    <span>Drive health</span>
+    {% if drives_at %}<span style="font-weight:400;color:var(--text-dim);font-size:0.72rem;">read {{ drives_at|ago }}</span>{% endif %}
+  </div>
+  {% if drives_error %}
+  <p class="field-hint" style="margin:0;">Not read: {{ drives_error }}.</p>
+  {% endif %}
+  {% for d in drives %}
+  {% set tone, label = drive_badge[d.state] %}
+  <div style="display:flex;justify-content:space-between;align-items:baseline;gap:0.75rem;font-size:0.8rem;padding:0.2rem 0;">
+    <span style="min-width:0;overflow:hidden;text-overflow:ellipsis;white-space:nowrap;" title="{{ d.model }}">
+      {{ status_badge(tone, label) }}
+      <span style="font-family:ui-monospace,monospace;">{{ d.device }}</span>
+      {% if d.model %}<span style="color:var(--text-dim);">{{ d.model }}{% if d.capacity_bytes %} · {{ d.capacity_bytes|bytes }}{% endif %}</span>{% endif %}
+      {% if d.why %}<span style="color:var(--text-muted);"> — {{ d.why }}</span>{% endif %}
+    </span>
+    <span style="color:var(--text-muted);font-variant-numeric:tabular-nums;white-space:nowrap;">
+      {%- set parts = [] -%}
+      {%- if d.get("life_used_pct") is not none %}{% set _ = parts.append("%d%% life used"|format(d.life_used_pct)) %}{% endif -%}
+      {%- if d.get("bad_sectors") is not none %}{% set _ = parts.append("%d bad sectors"|format(d.bad_sectors)) %}{% endif -%}
+      {%- if d.get("temp_c") is not none %}{% set _ = parts.append("%d°C"|format(d.temp_c)) %}{% endif -%}
+      {{ parts|join(" · ") if parts else "—" }}
+    </span>
+  </div>
+  {% endfor %}
+</div>
+{% endif %}
+
 {# ── Interfaces / disks (short panels, side by side at natural height) ─────── #}
 {% if nets or disks_io %}
 <div style="display:grid;grid-template-columns:repeat(auto-fit,minmax(260px,1fr));gap:1rem;align-items:start;">

plugins/host_agent/templates/install.sh.j2

@@ -63,6 +63,12 @@ chmod 0640 "$CONF_FILE"
 curl -fsSL "${FILES_URL}/maint.py" -o "$AGENT_DIR/maint.py"
 chmod 0755 "$AGENT_DIR/maint.py"
 chown root:root "$AGENT_DIR/maint.py"
+# Its packages (smartctl for drive health, #5789), with whichever package
+# manager this host has; one that can't be installed is said, never fatal.
+curl -fsSL "${FILES_URL}/install-deps.sh" -o "$AGENT_DIR/install-deps.sh"
+chmod 0755 "$AGENT_DIR/install-deps.sh"
+chown root:root "$AGENT_DIR/install-deps.sh"
+"$AGENT_DIR/install-deps.sh" || true
 install -d -o root -g root -m 0755 /var/lib/steward-agent /var/lib/steward-maint /var/lib/steward-maint/results
 install -d -o "$AGENT_USER" -g "$AGENT_USER" -m 0700 /var/lib/steward-agent/requests
 for u in $UNITS; do