Fix false recharge-cycle detection from a single noisy battery reading

A report was flagged as "the battery got recharged" (resetting
battery_history and stats_recharge_cycles, and re-arming the low-battery
alert) whenever it came in >= RECHARGE_JUMP_PCT above the single
immediately-previous report. That's exactly what a real recharge looks
like, but it's also exactly what a normal reading looks like right
after one noisy low report: e.g. 60, 59, 58, then a stray 53, then back
to a perfectly normal 58 -- 58 >= 53+5 falsely read as a recharge.

Now compared against the max of the last RECHARGE_LOOKBACK (3) reports
instead of just the one before it, so a lone stray reading doesn't get
to set the bar a normal reading then trips. A real recharge still needs
to clear all of them, so genuine recharges are still caught immediately
(verified: 18% -> 90% still triggers, history still resets).

Paired with the firmware-side battery.c change (trimmed-mean ADC
sampling) that reduces how often a stray reading like the 53 above
happens in the first place.

server/app/routers/device.py

@@ -28,6 +28,7 @@ from .common import (
     BATTERY_HISTORY_MAX,
     BATTERY_LOG_MAX,
     RECHARGE_JUMP_PCT,
+    RECHARGE_LOOKBACK,
     immich_client_for,
     immich_creds,
     list_assets,
@@ -202,7 +203,12 @@ def frame_battery(
     alert_frame_name = ""
     with frame_locked(db, frame.id) as locked:
         locked.stats_battery_reports += 1
-        if locked.battery_history and body.percent >= locked.battery_history[-1][1] + RECHARGE_JUMP_PCT:
+        # See RECHARGE_LOOKBACK: compared against the max of the last few
+        # reports, not just the single previous one, so a lone noisy dip
+        # can't make the next normal reading look like a recharge.
+        recent = locked.battery_history[-RECHARGE_LOOKBACK:]
+        recent_max = max((pct for _, pct in recent), default=None)
+        if recent_max is not None and body.percent >= recent_max + RECHARGE_JUMP_PCT:
             # Percent jumped up meaningfully -- the battery was recharged
             # (or swapped). Start a fresh discharge cycle so runtime and
             # discharge-rate estimates never span a charge -- and let a