fix(sway): only force a post-resume modeset when the stall is actually present

recover-output-after-resume.sh existed to work around the old ThinkPad's
/etc/modprobe.d/i915.conf, which forced PSR2 and the deepest display
C-states; the flip pending across suspend never completed, so every later
commit returned EBUSY and only a full modeset cleared it.

That file is gone and this laptop's iGPU is driven by `xe` (Panther Lake has
no i915 support at all), so the original cause cannot recur. The modeset is
now conditional on the user journal actually reporting "Atomic commit
failed"/"Page-flip failed" within the last 3 minutes.

Unconditional disable/enable + `shikanectl reload` on every resume was itself
a hazard: it is the exact sequence that re-triggers the shikane
VariantInProgress deadlock the watchdog exists to recover from.

hypridle.conf's comment no longer references the deleted i915.conf.

.config/hypr/hypridle.conf

@@ -1,34 +1,48 @@
+# Idle / lock for SWAY (wlroots), not Hyprland.
+#
+# Two things in the original Hyprland template silently did nothing under sway:
+#   * `loginctl lock-session`  -> sway implements no freedesktop login1 Session
+#                                 Lock protocol, so it is a no-op. Lock directly.
+#   * `hyprctl dispatch dpms`  -> "HYPRLAND_INSTANCE_SIGNATURE not set!".
+#                                 Use `swaymsg "output * power on|off"`.
+# Also: a hyprlock instance can outlive a sway session and then make the
+# `pidof hyprlock || hyprlock` guard a permanent no-op; the sway config kills
+# leftovers at login (`exec --no-startup-id pkill -x hyprlock`).
+
 general {
-    lock_cmd = pidof hyprlock || hyprlock       # avoid starting multiple hyprlock instances.
-    before_sleep_cmd = loginctl lock-session    # lock before suspend.
-    after_sleep_cmd = hyprctl dispatch dpms on  # to avoid having to press a key twice to turn on the display.
+    lock_cmd = pidof hyprlock || hyprlock
+    # pre-suspend uses swaylock: fast, sway-native, exits cleanly (no leak)
+    before_sleep_cmd = swaylock -f -c 000000
+    # DPMS "power on" alone does not clear a post-resume eDP page-flip stall
+    # ("Atomic commit failed: Device or resource busy" for tens of minutes).
+    # The script does a full disable/enable (forced modeset) but ONLY when the
+    # journal actually shows the stall -- it used to fire unconditionally, which
+    # risked the shikane VariantInProgress deadlock on healthy resumes. The
+    # original i915 PSR2/DC4 cause is gone; this laptop uses the `xe` driver.
+    # Lid-aware so it does not race shikane's "External Display" profile.
+    after_sleep_cmd = /home/work/.config/sway/scripts/recover-output-after-resume.sh
 }
 
 listener {
-    timeout = 150                                # 2.5min.
-    on-timeout = brightnessctl -s set 10         # set monitor backlight to minimum, avoid 0 on OLED monitor.
-    on-resume = brightnessctl -r                 # monitor backlight restore.
+    timeout = 150                                # 2.5min
+    on-timeout = brightnessctl -s set 10
+    on-resume = brightnessctl -r
 }
 
-# turn off keyboard backlight, comment out this section if you dont have a keyboard backlight.
-listener { 
-    timeout = 150                                          # 2.5min.
-    on-timeout = brightnessctl -sd rgb:kbd_backlight set 0 # turn off keyboard backlight.
-    on-resume = brightnessctl -rd rgb:kbd_backlight        # turn on keyboard backlight.
+# keyboard backlight
+listener {
+    timeout = 150
+    on-timeout = brightnessctl -sd rgb:kbd_backlight set 0
+    on-resume = brightnessctl -rd rgb:kbd_backlight
 }
 
 listener {
-    timeout = 300                                 # 5min
-    on-timeout = loginctl lock-session            # lock screen when timeout has passed
+    timeout = 300                                # 5min -> lock
+    on-timeout = pidof hyprlock || hyprlock
 }
 
 listener {
-    timeout = 330                                                     # 5.5min
-    on-timeout = hyprctl dispatch dpms off                            # screen off when timeout has passed
-    on-resume = hyprctl dispatch dpms on && brightnessctl -r          # screen on when activity is detected after timeout has fired.
-}
-
-listener {
-    timeout = 1800                                # 30min
-    on-timeout = systemctl suspend                # suspend pc
+    timeout = 330                                # 5.5min -> output power off
+    on-timeout = swaymsg "output * power off"
+    on-resume = swaymsg "output * power on" && brightnessctl -r
 }