Auto merge of rust-lang#87150 - rusticstuff:simplify_wrapping_neg, r=…

…m-ou-se

Make wrapping_neg() use wrapping_sub(), #[inline(always)]

This is a follow-up change to the fix for rust-lang#75598. It simplifies the implementation of wrapping_neg() for all integer types by just calling 0.wrapping_sub(self) and always inlines it. This leads to much less assembly code being emitted for opt-level≤1 and thus much better performance for debug-compiled code.

Background is [this discussion on the internals forum](https://internals.rust-lang.org/t/why-does-rust-generate-10x-as-much-unoptimized-assembly-as-gcc/14930).

library/core/src/num/int_macros.rs

@@ -1132,9 +1132,9 @@ macro_rules! int_impl {
         /// ```
         #[stable(feature = "num_wrapping", since = "1.2.0")]
         #[rustc_const_stable(feature = "const_int_methods", since = "1.32.0")]
-        #[inline]
+        #[inline(always)]
         pub const fn wrapping_neg(self) -> Self {
-            self.overflowing_neg().0
+            (0 as $SelfT).wrapping_sub(self)
         }
 
         /// Panic-free bitwise shift-left; yields `self << mask(rhs)`, where `mask` removes

library/core/src/num/uint_macros.rs

@@ -1246,9 +1246,9 @@ macro_rules! uint_impl {
         /// ```
         #[stable(feature = "num_wrapping", since = "1.2.0")]
         #[rustc_const_stable(feature = "const_wrapping_math", since = "1.32.0")]
-        #[inline]
+        #[inline(always)]
         pub const fn wrapping_neg(self) -> Self {
-            self.overflowing_neg().0
+            (0 as $SelfT).wrapping_sub(self)
         }
 
         /// Panic-free bitwise shift-left; yields `self << mask(rhs)`,