V4L2SVDA/VAAPIVDA: use visible size from decoder and pass to client

media/gpu/h264_decoder.h

Index: media/gpu/h264_decoder.h
diff --git a/media/gpu/h264_decoder.h b/media/gpu/h264_decoder.h
index 1caf662a7a6fa07b68b069321e014195fd6f28fd..a0daf6918455a46c4bdb4df4d86b2f2768291b0e 100644
--- a/media/gpu/h264_decoder.h
+++ b/media/gpu/h264_decoder.h
@@ -18,6 +18,7 @@
 #include "media/gpu/accelerated_video_decoder.h"
 #include "media/gpu/h264_dpb.h"
 #include "media/gpu/media_gpu_export.h"
+#include "ui/gfx/geometry/rect.h"
 #include "ui/gfx/geometry/size.h"
 
 namespace media {
@@ -266,6 +267,8 @@ class MEDIA_GPU_EXPORT H264Decoder : public AcceleratedVideoDecoder {
 
   // Output picture size.
   gfx::Size pic_size_;
+  // Output visible cropping rect.
+  gfx::Rect visible_rect_;
 
   // PicOrderCount of the previously outputted frame.
   int last_output_poc_;