[audio] Fix release buffer for audio input (#3336)

fixes microphone release buffer and data. Tested with Let's Sing 2025

Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3336
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Co-authored-by: Maufeat <sahyno1996@gmail.com>
Co-committed-by: Maufeat <sahyno1996@gmail.com>

src/audio_core/device/audio_buffers.h

@@ -113,6 +113,8 @@ public:
                 break;
             }
 
+            session.ReleaseBuffer(buffers[index]);
+
             ReleaseBuffer(index, core_timing.GetGlobalTimeNs().count());
             buffer_released = true;
         }

src/audio_core/sink/sink_stream.cpp

@@ -151,7 +151,17 @@ void SinkStream::ProcessAudioIn(std::span<const s16> input_buffer, std::size_t n
             playing_buffer.consumed = true;
     }
 
-    std::memcpy(&last_frame[0], &input_buffer[(frames_written - 1) * frame_size], frame_size_bytes);
+    if (frames_written > 0) {
+        std::memcpy(&last_frame[0], &input_buffer[(frames_written - 1) * frame_size], frame_size_bytes);
+    }
+
+    // update sample counts für audio-ins
+    {
+        std::scoped_lock lk{sample_count_lock};
+        last_sample_count_update_time = system.CoreTiming().GetGlobalTimeNs();
+        min_played_sample_count = max_played_sample_count;
+        max_played_sample_count += frames_written;
+    }
 }
 
 void SinkStream::ProcessAudioOutAndRender(std::span<s16> output_buffer, std::size_t num_frames) {