[libc++] Fix return type of ilogb(double) (#150374)

This commit addresses a seemingly unintentional return type of the ilogb
overload taking a double. Currently it returns double, while it is
supposed to return int.

The return types seems to be covered by libcxx/test/std/numerics/c.math/cmath.pass.cpp,
but the issue would only show up if we tested with a libc that doesn't
provide the ilogb(double) overload, which we don't.

Author: steffenlarsen

libcxx/include/__cxx03/__math/logarithms.h

@@ -58,7 +58,7 @@ inline _LIBCPP_HIDE_FROM_ABI double log10(_A1 __x) _NOEXCEPT {
 inline _LIBCPP_HIDE_FROM_ABI int ilogb(float __x) _NOEXCEPT { return __builtin_ilogbf(__x); }
 
 template <class = int>
-_LIBCPP_HIDE_FROM_ABI double ilogb(double __x) _NOEXCEPT {
+_LIBCPP_HIDE_FROM_ABI int ilogb(double __x) _NOEXCEPT {
   return __builtin_ilogb(__x);
 }
 

libcxx/include/__math/logarithms.h

@@ -58,7 +58,7 @@ inline _LIBCPP_HIDE_FROM_ABI double log10(_A1 __x) _NOEXCEPT {
 inline _LIBCPP_HIDE_FROM_ABI int ilogb(float __x) _NOEXCEPT { return __builtin_ilogbf(__x); }
 
 template <class = int>
-_LIBCPP_HIDE_FROM_ABI double ilogb(double __x) _NOEXCEPT {
+_LIBCPP_HIDE_FROM_ABI int ilogb(double __x) _NOEXCEPT {
   return __builtin_ilogb(__x);
 }