[ARM] Implement widening and narrowing integer moves, vmovl, vqmovn.

src/arm/assembler-arm.h

Index: src/arm/assembler-arm.h
diff --git a/src/arm/assembler-arm.h b/src/arm/assembler-arm.h
index a29fbff8af83b83c1cb0045f7633a44bcc823d16..b07934faf5ffc45c05e7d7e7312d2ea3da27f5aa 100644
--- a/src/arm/assembler-arm.h
+++ b/src/arm/assembler-arm.h
@@ -1321,7 +1321,10 @@ class Assembler : public AssemblerBase {
   void vst1(NeonSize size,
             const NeonListOperand& src,
             const NeonMemOperand& dst);
+  // dt represents the narrower type
   void vmovl(NeonDataType dt, QwNeonRegister dst, DwVfpRegister src);
+  // dt represents the narrower type.
+  void vqmovn(NeonDataType dt, DwVfpRegister dst, QwNeonRegister src);
 
   // Only unconditional core <-> scalar moves are currently supported.
   void vmov(NeonDataType dt, DwVfpRegister dst, int index, Register src);