Refactor Sk2x<T> + Sk4x<T> into SkNf<N,T> and SkNi<N,T>

src/opts/SkPMFloat_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.
  */
 
-inline SkPMFloat& SkPMFloat::operator=(const SkPMFloat& that) {
-    fColors = that.fColors;
-    return *this;
-}
-
 // For SkPMFloat(SkPMFColor), we widen our 8 bit components (fix8) to 8-bit components in 16 bits
 // (fix8_16), then widen those to 8-bit-in-32-bits (fix8_32), and finally convert those to floats.
 
 // get() and clamped() do the opposite, working from floats to 8-bit-in-32-bit,
 // to 8-bit-in-16-bit, back down to 8-bit components.
 // clamped() uses vqmovn to clamp while narrowing instead of just narrowing with vmovn.
 
 inline SkPMFloat::SkPMFloat(SkPMColor c) {
     SkPMColorAssert(c);
     uint8x8_t   fix8    = (uint8x8_t)vdup_n_u32(c);
     uint16x8_t  fix8_16 = vmovl_u8(fix8);
     uint32x4_t  fix8_32 = vmovl_u16(vget_low_u16(fix8_16));
     fColors = vcvtq_f32_u32(fix8_32);
     SkASSERT(this->isValid());
 }
 
 inline SkPMColor SkPMFloat::trunc() const {
-    uint32x4_t  fix8_32  = vcvtq_u32_f32(fColors);  // vcvtq_u32_f32 truncates
+    uint32x4_t  fix8_32  = vcvtq_u32_f32(fColors.vec());  // vcvtq_u32_f32 truncates
     uint16x4_t  fix8_16  = vmovn_u32(fix8_32);
     uint8x8_t   fix8     = vmovn_u16(vcombine_u16(fix8_16, vdup_n_u16(0)));
     SkPMColor c = vget_lane_u32((uint32x2_t)fix8, 0);
     SkPMColorAssert(c);
     return c;
 }
 
 inline SkPMColor SkPMFloat::get() const {
     SkASSERT(this->isValid());
-    return SkPMFloat(Sk4f(0.5f) + *this).trunc();
+    return SkPMFloat(Sk4s(0.5f) + *this).trunc();
 }
 
 inline SkPMColor SkPMFloat::clamped() const {
-    float32x4_t add_half = vaddq_f32(fColors, vdupq_n_f32(0.5f));
+    float32x4_t add_half = vaddq_f32(fColors.vec(), vdupq_n_f32(0.5f));
     uint32x4_t  fix8_32  = vcvtq_u32_f32(add_half);  // vcvtq_u32_f32 truncates, so round manually
     uint16x4_t  fix8_16  = vqmovn_u32(fix8_32);
     uint8x8_t   fix8     = vqmovn_u16(vcombine_u16(fix8_16, vdup_n_u16(0)));
     SkPMColor c = vget_lane_u32((uint32x2_t)fix8, 0);
     SkPMColorAssert(c);
     return c;
 }
 
 // TODO: we should be able to beat these loops on all three methods.
 inline void SkPMFloat::From4PMColors(const SkPMColor colors[4],
@@ skipping 14 matching lines @@
 }
 
 inline void SkPMFloat::ClampTo4PMColors(
         const SkPMFloat& a, const SkPMFloat& b, const SkPMFloat&c, const SkPMFloat& d,
         SkPMColor colors[4]) {
     colors[0] = a.clamped();
     colors[1] = b.clamped();
     colors[2] = c.clamped();
     colors[3] = d.clamped();
 }