fix(recognition): play stop sound after audio thread joins to prevent capture on short recordings (#436)

Moves play_stop_sound() to after audio_thread.join() but before the
final buffer processing. This closes the race where the audio thread's
in-flight chunk could capture the stop cue for short (≤ ~192 ms) recordings,
while still feeling immediate compared to the original placement after all
recognition cleanup.

src/vocalinux/speech_recognition/recognition_manager.py

@@ -1768,16 +1768,16 @@ class SpeechRecognitionManager:
         # Stop recording FIRST to prevent capturing the stop sound
         self.should_record = False
 
-        # Give immediate release feedback. The recorder may still finish one
-        # in-flight chunk, so the stop-sound guard below keeps the cue out.
-        play_stop_sound()
-
         # Wait for audio thread to finish recording and enqueue any pending audio
         # This is critical to prevent race condition where recognition thread exits
         # before the final audio segment is enqueued
         if self.audio_thread and self.audio_thread.is_alive():
             self.audio_thread.join(timeout=2.0)
 
+        # Play stop sound now that the audio thread is done and cannot capture it.
+        # Kept before buffer processing so the cue still feels immediate.
+        play_stop_sound()
+
         # Trim only a small tail to avoid the stop sound without clipping the user's final word.
         with self._buffer_lock:
             stop_sound_guard_chunks = self._get_stop_sound_guard_chunks()