fleet-manager: chrony tracking metrics relative to the NTP source

The agent only reported chrony as synced|unsynced — nothing about how well the
clock tracks. Parse `chronyc tracking` for the source-relative metrics.

- proto: HealthSnapshot.chrony_offset_seconds (16, signed system-time offset,
  fast=+), chrony_source (17, reference name/IP), chrony_stratum (18); regenerated.
- agent: replace chronyState() with chronyTracking()/parseChrony() — one exec,
  parses Leap status + Reference ID + Stratum + System time offset (+ test).
- manager: Health.{ChronyOffset,ChronySource,ChronyStratum}; snapshotToHealth maps
  them; influx writes chrony_offset_seconds/_stratum/_source and whitelists the
  offset + stratum as graphable fields.
- UI: chronySummary() + offsetMs() — the chrony row now shows
  "synced · 192.168.50.10 · s3 · +0.01ms" instead of just "synced".

serving/fleet-manager/ui/src/pages/Monitor.svelte

@@ -3,7 +3,7 @@
   import Panel from '../components/Panel.svelte'
   import StatePill from '../components/StatePill.svelte'
   import Chart from '../components/Chart.svelte'
-  import { relTime, pct, num, healthState, shortSerial, dur, rebootStatus } from '../lib/fmt.js'
+  import { relTime, pct, num, healthState, shortSerial, dur, rebootStatus, chronySummary } from '../lib/fmt.js'
 
   let hosts = $state([])
   let error = $state('')
@@ -103,7 +103,7 @@
           <dt class="text-faint">auth</dt><dd class="text-muted">{selected.health?.auth || '—'}</dd>
           <dt class="text-faint">nvme root</dt><dd class="text-muted">{selected.health?.nvme_root || '—'}</dd>
           <dt class="text-faint">selinux</dt><dd class="text-muted">{selected.health?.selinux || '—'}</dd>
-          <dt class="text-faint">chrony</dt><dd class="text-muted">{selected.health?.chrony || '—'}</dd>
+          <dt class="text-faint">chrony</dt><dd class="text-muted">{chronySummary(selected.health)}</dd>
           <dt class="text-faint">kiosk</dt><dd class="text-muted">{selected.health?.kiosk_active ? 'active' : 'idle'} · {selected.usage?.session_uptime || '—'}</dd>
           <dt class="text-faint">uptime</dt><dd class="text-muted">{selected.health?.uptime_seconds ? dur(selected.health.uptime_seconds) : '—'}</dd>
           {#if rebootStatus(selected)}