Fix iOS background audio pacing

Sleep-paced threads stall for hundreds of milliseconds when iOS coalesces a
backgrounded app's wakeups, and the deadline resets that discarded the deficit
left the capture backlog queued until its ring overflowed: regular dropouts
that worsen the longer the app stays backgrounded. Pace both 20 ms hands-offs
from the AVAudioSourceNode render callback instead (mix via Audio.RunCycle
with deviceClockedAudio, capture handoff from the ring), wake the UDP sender
on a queue signal instead of a 1 ms poll, and let stalled consumers drop their
backlog to the buffer target instead of ratcheting it.

src/VoiceCat.Core/VoiceCatClient.cs

@@ -55,12 +55,15 @@ public sealed partial class VoiceCatClient : IAsyncDisposable
     public bool TryReadEvent(out Envelope? envelope) => events.Reader.TryRead(out envelope);
     public IAsyncEnumerable<Envelope> ReadEventsAsync(CancellationToken cancellationToken = default) => events.Reader.ReadAllAsync(cancellationToken);
 
-    public VoiceCatClient(string clientName = "VoiceCat .NET", string clientVersion = "0.1.0", string? tofuStorePath = null)
+    // deviceClockedAudio: the platform's audio callback drives Audio.RunCycle() itself (iOS
+    // keeps render callbacks running in the background while sleep-paced threads coalesce there),
+    // so the audio engine must not start its own pacing worker.
+    public VoiceCatClient(string clientName = "VoiceCat .NET", string clientVersion = "0.1.0", string? tofuStorePath = null, bool deviceClockedAudio = false)
     {
         this.clientName = clientName;
         this.clientVersion = clientVersion;
         pins = new(tofuStorePath ?? Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.LocalApplicationData), "VoiceCat", "tofu.txt"));
-        Audio = new(TrySendEncodedVoice);
+        Audio = new(TrySendEncodedVoice, startWorker: !deviceClockedAudio);
         VoiceReceived += Audio.Receive;
     }