Normative: Allow NaN values to be *optionally* canoncalized

This patch legalizes V8's occasional canonicalization of NaNs
by changing SetValueInBuffer to *either* a particular value
for the implementation-distinguishable NaN value *or* an
implementation-defined canonical value.

This semantic change reached consensus at the November 2016
TC39 meeting.

Closes tc39#635

spec.html

@@ -35819,9 +35819,9 @@ <h1>NumberToRawBytes( _type_, _value_, _isLittleEndian_ )</h1>
         <p>The abstract operation NumberToRawBytes takes three parameters, a String _type_, a Number _value_, and a Boolean _isLittleEndian_. This operation performs the following steps:</p>
         <emu-alg>
           1. If _type_ is `"Float32"`, then
-            1. Let _rawBytes_ be a List containing the 4 bytes that are the result of converting _value_ to IEEE 754-2008 binary32 format using &ldquo;Round to nearest, ties to even&rdquo; rounding mode. If _isLittleEndian_ is *false*, the bytes are arranged in big endian order. Otherwise, the bytes are arranged in little endian order. If _value_ is *NaN*, _rawBytes_ may be set to any implementation chosen IEEE 754-2008 binary32 format Not-a-Number encoding. An implementation must always choose the same encoding for each implementation distinguishable *NaN* value.
+            1. Let _rawBytes_ be a List containing the 4 bytes that are the result of converting _value_ to IEEE 754-2008 binary32 format using &ldquo;Round to nearest, ties to even&rdquo; rounding mode. If _isLittleEndian_ is *false*, the bytes are arranged in big endian order. Otherwise, the bytes are arranged in little endian order. If _value_ is *NaN*, _rawBytes_ may be set to any implementation chosen IEEE 754-2008 binary32 format Not-a-Number encoding. An implementation must always choose the same encoding for each implementation distinguishable *NaN* value, or an implementation-defined canonical value.
           1. Else if _type_ is `"Float64"`, then
-            1. Let _rawBytes_ be a List containing the 8 bytes that are the IEEE 754-2008 binary64 format encoding of _value_. If _isLittleEndian_ is *false*, the bytes are arranged in big endian order. Otherwise, the bytes are arranged in little endian order. If _value_ is *NaN*, _rawBytes_ may be set to any implementation chosen IEEE 754-2008 binary64 format Not-a-Number encoding. An implementation must always choose the same encoding for each implementation distinguishable *NaN* value.
+            1. Let _rawBytes_ be a List containing the 8 bytes that are the IEEE 754-2008 binary64 format encoding of _value_. If _isLittleEndian_ is *false*, the bytes are arranged in big endian order. Otherwise, the bytes are arranged in little endian order. If _value_ is *NaN*, _rawBytes_ may be set to any implementation chosen IEEE 754-2008 binary64 format Not-a-Number encoding. An implementation must always choose the same encoding for each implementation distinguishable *NaN* value, or an implementation-defined canonical value.
           1. Else,
             1. Let _n_ be the Number value of the Element Size specified in <emu-xref href="#table-49"></emu-xref> for Element Type _type_.
             1. Let _convOp_ be the abstract operation named in the Conversion Operation column in <emu-xref href="#table-49"></emu-xref> for Element Type _type_.