Change NetEq::GetAudio to use AudioFrame

With this change, NetEq now uses AudioFrame as output type, like the
surrounding functions in ACM and VoiceEngine already do.

The computational savings is probably slim, since one memcpy is
removed while another one is added (both in AcmReceiver::GetAudio).

More simplifications and clean-up will be done in
AcmReceiver::GetAudio in future CLs.

BUG=webrtc:5607

Review URL: https://codereview.webrtc.org/1750353002

Cr-Commit-Position: refs/heads/master@{#11874}

webrtc/modules/audio_coding/neteq/sync_buffer.h

@@ -13,6 +13,7 @@
 
 #include "webrtc/base/constructormagic.h"
 #include "webrtc/modules/audio_coding/neteq/audio_multi_vector.h"
+#include "webrtc/modules/include/module_common_types.h"
 #include "webrtc/typedefs.h"
 
 namespace webrtc {
@@ -65,8 +66,10 @@ class SyncBuffer : public AudioMultiVector {
 
   // Reads |requested_len| samples from each channel and writes them interleaved
   // into |output|. The |next_index_| is updated to point to the sample to read
-  // next time.
-  size_t GetNextAudioInterleaved(size_t requested_len, int16_t* output);
+  // next time. The AudioFrame |output| is first reset, and the |data_|,
+  // |interleaved_|, |num_channels_|, and |samples_per_channel_| fields are
+  // updated.
+  void GetNextAudioInterleaved(size_t requested_len, AudioFrame* output);
 
   // Adds |increment| to |end_timestamp_|.
   void IncreaseEndTimestamp(uint32_t increment);