Lower intrinsics::offset to mir::BinOp::Offset

They're semantically the same, so this means the backends don't need to handle the intrinsic and means fewer MIR basic blocks in pointer arithmetic code.

compiler/rustc_codegen_cranelift/src/intrinsics/mod.rs

@@ -534,7 +534,7 @@ fn codegen_regular_intrinsic_call<'tcx>(
 
         // The only difference between offset and arith_offset is regarding UB. Because Cranelift
         // doesn't have UB both are codegen'ed the same way
-        sym::offset | sym::arith_offset => {
+        sym::arith_offset => {
             intrinsic_args!(fx, args => (base, offset); intrinsic);
             let offset = offset.load_scalar(fx);