lavu: Remove bit packing from AVComponentDescriptor

There is no practical benefit in having this structure elements
bit packed given the size of the structure and its usage.
Change types from uint16_t (packed) to plain int in order to simplify
modifying the structure and accessing its fields.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>

doc/APIchanges

@@ -15,6 +15,8 @@ API changes, most recent first:
 
 2015-xx-xx - lavu 55.0.0
   xxxxxxx - Change type of AVPixFmtDescriptor.flags from uint8_t to uint64_t.
+  xxxxxxx - Change type of AVComponentDescriptor fields from uint16_t to int
+            and drop bit packing.
 
 2015-xx-xx - lavu 54.17.0
   xxxxxxx -  Add av_blowfish_alloc().

libavutil/pixdesc.h

@@ -31,30 +31,30 @@ typedef struct AVComponentDescriptor {
     /**
      * Which of the 4 planes contains the component.
      */
-    uint16_t plane        : 2;
+    int plane;
 
     /**
      * Number of elements between 2 horizontally consecutive pixels minus 1.
      * Elements are bits for bitstream formats, bytes otherwise.
      */
-    uint16_t step_minus1  : 3;
+    int step_minus1;
 
     /**
      * Number of elements before the component of the first pixel plus 1.
      * Elements are bits for bitstream formats, bytes otherwise.
      */
-    uint16_t offset_plus1 : 3;
+    int offset_plus1;
 
     /**
      * Number of least significant bits that must be shifted away
      * to get the value.
      */
-    uint16_t shift        : 3;
+    int shift;
 
     /**
      * Number of bits in the component minus 1.
      */
-    uint16_t depth_minus1 : 4;
+    int depth_minus1;
 } AVComponentDescriptor;
 
 /**