avcodec/mpegvideo: Don't overallocate buffer

Only encoders need two sets of int16_t [12][64]
(one to save the current best state and one for the current
working state); decoders need only one. This saves 1.5KiB
per slice context for a decoder.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

libavcodec/mpegvideo.c

@@ -381,7 +381,7 @@ static int init_duplicate_context(MpegEncContext *s)
                 return AVERROR(ENOMEM);
         }
     }
-    if (!FF_ALLOCZ_TYPED_ARRAY(s->blocks,  2))
+    if (!FF_ALLOCZ_TYPED_ARRAY(s->blocks,  1 + s->encoding))
         return AVERROR(ENOMEM);
     s->block = s->blocks[0];