feat: add sidebar update indicator toggle and Stack Health badge (#1570)

* feat: add sidebar update indicator toggle and Stack Health badge

- Add image_update_sidebar_indicators setting (default off, node-scoped)
- Gate the Updates filter chip and sidebar status indicators on the setting
- Add "Update available" badge to Stack Health table (always visible)
- Extend ImageUpdateStatus with sidebarIndicators boolean
- Poll /api/image-updates/status alongside /detail in useImageUpdates
- React to SENCHO_SETTINGS_CHANGED for instant toggle propagation
- Reset sidebar state on node switch; generation-guard stale responses
- Disable toggle when status is null (loading) or field is absent (old node)
- Wire stackUpdates through ViewRouter → HomeDashboard → StackHealthTable
- Update settings registry, operator docs, and sidebar/dashboard docs

* fix: guard against stale node renders, memo drift, and cross-node error toasts

- Track owning node ID in useImageUpdates state so React never renders
  node B with node A's data before the passive effect resets (P2)
- Replace incorrect stackUpdates dependency with sidebarStackUpdates in
  chipFilteredFiles useMemo (P3)
- Guard the error toast in handleSidebarIndicatorsChange so a stale PATCH
  failure from node A does not surface while viewing node B (P3)

* fix: default sidebar update indicators to on (opt-out)

The sidebar indicators are a safe convenience that most users want.
Switching the default from off to on matches the opt-out convention
used by prune_on_update, reclaim_hero, and health_gate_enabled.

docs/features/dashboard.mdx

@@ -62,7 +62,7 @@ A mono table of every stack discovered in the active node's `COMPOSE_DIR`, sorte
 | Column | Description |
 |--------|-------------|
 | **Status dot** | Green when the stack is running and its 10-minute peak CPU is under 80%, amber when peak CPU is at or above 80%, rose when any container has exited or peak CPU is at or above 90% |
-| **STACK** | Stack name, derived from the compose file (extension stripped) |
+| **STACK** | Stack name with an orange "Update available" badge when a newer image has been detected. The badge appears regardless of the sidebar indicator setting. |
 | **HOST** | Active node this stack belongs to |
 | **UP** | How long the oldest running container has been up, in compact units (`s` / `m` / `h` / `d`); a stopped or never-started stack reads `--` |
 | **CPU** | Latest aggregate CPU across the stack's containers |

docs/features/sidebar.mdx

@@ -28,7 +28,7 @@ Four chips sit below the search box. Each shows a live count to the right of its
 - **All**: every stack on the node.
 - **Up**: stacks that are running with nothing crashed. A stack whose only stopped container finished cleanly (an init job that exited without error) still counts as up.
 - **Down**: stacks that need attention, whether fully stopped or running with at least one crashed container (the `PT` state described below).
-- **Updates**: stacks with at least one image update available. The chip renders in orange when the count is non-zero so you can spot pending updates at a glance.
+- **Updates**: stacks with at least one image update available. The chip renders in orange when the count is non-zero so you can spot pending updates at a glance. The Updates chip and the trailing update indicators on stack rows are controlled by the Image Update Checks [sidebar setting](/reference/settings#image-update-checks---sidebar). When the setting is off, the chip and indicators are hidden.
 
 <Frame>
   <img src="/images/sidebar/sidebar-filter-chips.png" alt="Filter chip row showing ALL (15), Up (15), Down (0), and Updates (1) with the Updates chip highlighted in orange and a collapse toggle icon on the right" />