fix: keep panel restarts from hanging on the CPU job and Telegram (#7029)

Restarting the panel left it unreachable for up to a minute. CheckCpuJob sampled cpu.Percent over its whole one-minute period, so cron.Stop nearly always waited on it; it now diffs cpu.Times against its previous run (iowait counted as idle, as before). With Telegram unreachable, bus.Stop waited on an in-flight send and tgbot.Start ran getMe/setMyCommands inline; each bot call now returns once StopBot closes the bot's stop channel, the bot stops before the bus, and the artwork/commands warm-up runs in the background.

internal/web/service/server.go

@@ -880,36 +880,12 @@ func (s *ServerService) sampleCPUUtilization() (float64, error) {
 		return s.emaCPU, nil
 	}
 
-	// Compute busy and total deltas
-	// Note: Guest and GuestNice times are already included in User and Nice respectively,
-	// so we exclude them to avoid double-counting (Linux kernel accounting)
-	idleDelta := cur.Idle - s.lastCPUTimes.Idle
-	busyDelta := (cur.User - s.lastCPUTimes.User) +
-		(cur.System - s.lastCPUTimes.System) +
-		(cur.Nice - s.lastCPUTimes.Nice) +
-		(cur.Iowait - s.lastCPUTimes.Iowait) +
-		(cur.Irq - s.lastCPUTimes.Irq) +
-		(cur.Softirq - s.lastCPUTimes.Softirq) +
-		(cur.Steal - s.lastCPUTimes.Steal)
-
-	totalDelta := busyDelta + idleDelta
-
-	// Update last sample for next time
+	raw, ok := sys.CPUBusyPercent(s.lastCPUTimes, cur)
 	s.lastCPUTimes = cur
-
-	// Guard against division by zero or negative deltas (e.g., counter resets)
-	if totalDelta <= 0 {
+	if !ok {
 		return s.emaCPU, nil
 	}
 
-	raw := 100.0 * (busyDelta / totalDelta)
-	if raw < 0 {
-		raw = 0
-	}
-	if raw > 100 {
-		raw = 100
-	}
-
 	// Exponential moving average to smooth spikes
 	const alpha = 0.3 // smoothing factor (0<alpha<=1). Higher = more responsive, lower = smoother
 	if s.emaCPU == 0 {