specialize for std::float32_t and std::float64_t explicitly

credit: @lemire

Author: lemire

include/fast_float/float_common.h

@@ -1170,6 +1170,9 @@ static_assert(std::is_same<equiv_uint_t<std::float64_t>, uint64_t>::value,
 static_assert(
     std::numeric_limits<std::float64_t>::is_iec559,
     "std::float64_t must fulfill the requirements of IEC 559 (IEEE 754)");
+
+template <>
+struct binary_format<std::float64_t> : public binary_format<double> {};
 #endif // __STDCPP_FLOAT64_T__
 
 #ifdef __STDCPP_FLOAT32_T__
@@ -1178,6 +1181,9 @@ static_assert(std::is_same<equiv_uint_t<std::float32_t>, uint32_t>::value,
 static_assert(
     std::numeric_limits<std::float32_t>::is_iec559,
     "std::float32_t must fulfill the requirements of IEC 559 (IEEE 754)");
+
+template <>
+struct binary_format<std::float32_t> : public binary_format<float> {};
 #endif // __STDCPP_FLOAT32_T__
 
 #ifdef __STDCPP_FLOAT16_T__
@@ -1249,7 +1255,6 @@ constexpr chars_format adjust_for_feature_macros(chars_format fmt) {
       ;
 }
 } // namespace detail
-
 } // namespace fast_float
 
 #endif