[libc] Use __builtin_ctzll instead of __builtin_ctzl in elements_x86.h.

__builtin_ctzl takes an unsigned long argument which need not be 64-bit
long on all platforms. Using __builtin_ctzll, which takes an unsigned
long long argument, ensures that 64-bit values will be handled on a
wider range of platforms.

Without this change, the test corresponding to M512 fails in Windows.

Reviewed By: gchatelet

Differential Revision: https://reviews.llvm.org/D104897

libc/src/string/memory_utils/elements_x86.h

@@ -46,7 +46,7 @@ template <typename Base> struct Vector : public Base {
   }
 
   static int CharDiff(const char *a, const char *b, uint64_t mask) {
-    const size_t diff_index = __builtin_ctzl(mask);
+    const size_t diff_index = __builtin_ctzll(mask);
     const int ca = (unsigned char)a[diff_index];
     const int cb = (unsigned char)b[diff_index];
     return ca - cb;