Replace our home-grown BitCast with bit_cast from Chrome/Google3.

src/double.h

 // Copyright 2011 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #ifndef V8_DOUBLE_H_
 #define V8_DOUBLE_H_
 
 #include "src/diy-fp.h"
 
 namespace v8 {
 namespace internal {
 
 // We assume that doubles and uint64_t have the same endianness.
-inline uint64_t double_to_uint64(double d) { return BitCast<uint64_t>(d); }
-inline double uint64_to_double(uint64_t d64) { return BitCast<double>(d64); }
+inline uint64_t double_to_uint64(double d) { return bit_cast<uint64_t>(d); }
+inline double uint64_to_double(uint64_t d64) { return bit_cast<double>(d64); }
 
 // Helper functions for doubles.
 class Double {
  public:
   static const uint64_t kSignMask = V8_2PART_UINT64_C(0x80000000, 00000000);
   static const uint64_t kExponentMask = V8_2PART_UINT64_C(0x7FF00000, 00000000);
   static const uint64_t kSignificandMask =
       V8_2PART_UINT64_C(0x000FFFFF, FFFFFFFF);
   static const uint64_t kHiddenBit = V8_2PART_UINT64_C(0x00100000, 00000000);
   static const int kPhysicalSignificandSize = 52;  // Excludes the hidden bit.
@@ skipping 174 matching lines @@
       biased_exponent = static_cast<uint64_t>(exponent + kExponentBias);
     }
     return (significand & kSignificandMask) |
         (biased_exponent << kPhysicalSignificandSize);
   }
 };
 
 } }  // namespace v8::internal
 
 #endif  // V8_DOUBLE_H_