[libc] Include nextafter's functions to Windows's entrypoints

[hedingarcia]

Summary

Incorporated the varied functions for nextafter and refactored
NextAfterTest.h to correctly define bitWidthOfType for both
Linux and Windows; by letting FloatProperties take care
of the directives' logic based on the platform being used.
This allows to successfully run nextafter's tests.

libc/config/windows/entrypoints.txt

set(TARGET_LIBM_ENTRYPOINTS
     libc.src.math.ldexpf
     libc.src.math.ldexpl
     libc.src.math.logb
     libc.src.math.logbf
     libc.src.math.logbl
     libc.src.math.modf
     libc.src.math.modff
     libc.src.math.modfl
+    libc.src.math.nextafter
+    libc.src.math.nextafterf
+    libc.src.math.nextafterl
 )
 
 set(TARGET_LLVMLIBC_ENTRYPOINTS
   ${TARGET_LIBC_ENTRYPOINTS}
   ${TARGET_LIBM_ENTRYPOINTS}
 )

libc/test/src/math/NextAfterTest.h

 #include <math.h>
 
 template <typename T>
 class NextAfterTestTemplate : public __llvm_libc::testing::Test {
   using FPBits = __llvm_libc::fputil::FPBits<T>;
   using MantissaWidth = __llvm_libc::fputil::MantissaWidth<T>;
   using UIntType = typename FPBits::UIntType;
 
-#if (defined(__x86_64__) || defined(__i386__))
   static constexpr int bitWidthOfType =
-      __llvm_libc::cpp::IsSame<T, long double>::Value ? 80 : (sizeof(T) * 8);
-#else
-  static constexpr int bitWidthOfType = sizeof(T) * 8;
-#endif
+      __llvm_libc::fputil::FloatProperties<T>::bitWidth;
 
   const T zero = T(FPBits::zero());
   const T negZero = T(FPBits::negZero());
   const T inf = T(FPBits::inf());
   const T negInf = T(FPBits::negInf());
   const T nan = T(FPBits::buildNaN(1));
   const UIntType minSubnormal = FPBits::minSubnormal;
   const UIntType maxSubnormal = FPBits::maxSubnormal;