YJIT: x64: Remove register shuffle with opt_and and friends (#10498)

This is best understood by looking at the change to the output:

```diff
  # Insn: 0002 opt_and (stack_size: 2)
  - mov rax, rsi
  - and rax, rdi
  - mov rsi, rax
  + and rsi, rdi
```

It's a bit awkward to match against due to how stack operands are
lowered, but hey, it's nice to save the 2 unnecessary MOVs.

yjit/src/backend/x86_64/mod.rs

@@ -181,6 +181,23 @@ impl Assembler
                             iterator.map_insn_index(&mut asm);
                             iterator.next_unmapped(); // Pop merged Insn::Mov
                         }
+                        (Opnd::Reg(_), Opnd::Reg(_), Some(Insn::Mov { dest, src }))
+                        if out == src && live_ranges[index] == index + 1 && {
+                            // We want to do `dest == left`, but `left` has already gone
+                            // through lower_stack_opnd() while `dest` has not. So we
+                            // lower `dest` before comparing.
+                            let lowered_dest = if let Opnd::Stack { .. } = dest {
+                                asm.lower_stack_opnd(dest)
+                            } else {
+                                *dest
+                            };
+                            lowered_dest == *left
+                        } => {
+                            *out = *dest;
+                            asm.push_insn(insn);
+                            iterator.map_insn_index(&mut asm);
+                            iterator.next_unmapped(); // Pop merged Insn::Mov
+                        }
                         _ => {
                             match (unmapped_opnds[0], unmapped_opnds[1]) {
                                 (Opnd::Mem(_), Opnd::Mem(_)) => {

yjit/src/codegen.rs

@@ -3827,7 +3827,7 @@ fn gen_opt_and(
 
         // Push the output on the stack
         let dst = asm.stack_push(Type::Fixnum);
-        asm.store(dst, val);
+        asm.mov(dst, val);
 
         Some(KeepCompiling)
     } else {
@@ -3867,7 +3867,7 @@ fn gen_opt_or(
 
         // Push the output on the stack
         let dst = asm.stack_push(Type::Fixnum);
-        asm.store(dst, val);
+        asm.mov(dst, val);
 
         Some(KeepCompiling)
     } else {
@@ -3909,7 +3909,7 @@ fn gen_opt_minus(
 
         // Push the output on the stack
         let dst = asm.stack_push(Type::Fixnum);
-        asm.store(dst, val);
+        asm.mov(dst, val);
 
         Some(KeepCompiling)
     } else {