Shrink TyKind::FnPtr.

By splitting the `FnSig` within `TyKind::FnPtr` into `FnSigTys` and
`FnHeader`, which can be packed more efficiently. This reduces the size
of the hot `TyKind` type from 32 bytes to 24 bytes on 64-bit platforms.
This reduces peak memory usage by a few percent on some benchmarks. It
also reduces cache misses and page faults similarly, though this doesn't
translate to clear cycles or wall-time improvements on CI.

compiler/rustc_hir_analysis/src/coherence/inherent_impls.rs

@@ -172,7 +172,7 @@ impl<'tcx> InherentCollect<'tcx> {
             | ty::RawPtr(_, _)
             | ty::Ref(..)
             | ty::Never
-            | ty::FnPtr(_)
+            | ty::FnPtr(..)
             | ty::Tuple(..) => self.check_primitive_impl(id, self_ty),
             ty::Alias(ty::Projection | ty::Inherent | ty::Opaque, _) | ty::Param(_) => {
                 Err(self.tcx.dcx().emit_err(errors::InherentNominal { span: item_span }))