### System details

Device: HP Spectre x360 2-in-1 Laptop 14-ef1xxx    Omarchy: 4.0.2-1    Kernel: 7.1.9-arch1-2   GPU: Intel Iris Xe (ADL GT2), i915     Sleep: s2idle only     Displays: internal eDP-1 only, no external monitor attached

### What's wrong?

On a laptop with no external monitor, closing and reopening the lid leaves the internal panel dark. The machine has resumed correctly and is fully responsive, but nothing lights the screen. It looks exactly like a failed resume, and every occurrence here was ended with a forced power-off.

Pressing any key recovers it. The power button never does, which is what makes it look like a hang rather than a blank screen.

### Why it happens, in shipped code

1. `lock/Service.qml`'s `idleBlankTimer` fires **5 seconds** after locking and runs `omarchy-brightness-display off`, which dispatches `dpms disable`. On lid close the sequence is: lock, five seconds later panel off, then suspend. The timer's own comment says fingerprint auth must not hold the panel up, so this happens on every lid-close lock.
2. Only two code paths ever dispatch `dpms enable`:
    - `omarchy-system-wake`, reached solely from `Keys.onPressed` on the lock screen's password field.
    - `dpms_internal enable` inside `omarchy-hyprland-monitor-clamshell`'s `enable_internal()`, which sits behind `if (( changed ))` — true only when a clamshell flag file was just removed.
3. **That flag is never written when there is no external monitor.** A laptop-only lid-open therefore runs the reconciler and dispatches nothing.
4. The power button cannot rescue it: logind consumes it and `10-ignore-power-button.conf` ignores it, so the compositor never sees a keypress.

### Evidence

A clean 70-minute resume, journal showing no fault at all:

```
14:08:07  Lid opened
14:08:07  PM: suspend exit
14:08:07  backlight intel_backlight: PM: backlight_resume returned 0
14:08:07  tailscaled: monitor: time jump detected (slept 1h10m12s)
   ... lock screen loops fingerprint every 10s, six times, panel dark ...
14:08:49  systemd-logind: Power key pressed short
14:08:55  systemd-logind: Power key pressed short
14:09:03  systemd-logind: Power key pressed short
14:09:05  <log ends mid-line>
```

Measured directly:

```
dpms before:      [true]
after blank:      [false]   <- omarchy-brightness-display off, i.e. what the lock screen runs
after wake:       [true]    <- omarchy-system-wake
```

### Why this is easy to miss

It looks fine whenever the first thing you do after opening the lid is type. It looks like a dead machine whenever you reach for the power button. That intermittency is what kept it unexplained here for two days.

### Local workaround

Replacing the default lid-open bind rather than adding a second one, so the reconciler still runs exactly once:

```lua
hl.unbind("switch:off:Lid Switch")
o.bind("switch:off:Lid Switch", "Lid open: reconcile displays, wake panel",
  "bash -c 'omarchy-hyprland-monitor-clamshell; omarchy-system-wake'",
  { locked = true })
```

### Suggested fix

Either dispatch `dpms enable` unconditionally on lid-open rather than only when the clamshell flag changed, or move `enable_internal()`'s DPMS step outside the `if (( changed ))` guard.