use new shuffle to speed up affine matrix mappts

src/opts/SkNx_neon.h

 /*
  * Copyright 2015 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */
 
 #ifndef SkNx_neon_DEFINED
 #define SkNx_neon_DEFINED
 
@@ skipping 378 matching lines @@
         return vbslq_u8(fVec, t.fVec, e.fVec);
     }
 
     uint8x16_t fVec;
 };
 
 #undef SHIFT32
 #undef SHIFT16
 #undef SHIFT8
 
+template <>
+inline SkNf<4,float> SkNx_shuffle_impl<SkNf<4,float>, 1,0,3,2>(const SkNf<4,float>& src) {
+    return vrev64q_f32(src.fVec);
+}
+
 }  // namespace
 
 #endif//SkNx_neon_DEFINED