Stop using to_hir_binop in codegen

[scottmcm]

This came up in #125359 (comment) , and looking into it we can just use the mir::BinOps directly instead of hir::BinOpKinds.

(AKA rather than going mir::BinOp → hir::BinOpKind → IntPredicate, just go mir::BinOp → IntPredicate.)

[rustbot]

r? @wesleywiser

rustbot has assigned @wesleywiser.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[RalfJung]

So where is that function even still used now? Maybe we can just get rid of it entirely...

[scottmcm]

It's used here

rust/compiler/rustc_middle/src/mir/terminator.rs

Lines 360 to 364 in b54dd08

	Overflow(binop, left, right) => {
	add!("op", binop.to_hir_binop().as_str());
	add!("left", format!("{left:#?}"));
	add!("right", format!("{right:#?}"));
	}

and

rust/compiler/rustc_middle/src/ty/print/pretty.rs

Lines 1537 to 1543 in b54dd08

	Expr::Binop(op, c1, c2) => {
	let precedence = |binop: rustc_middle::mir::BinOp| {
	use rustc_ast::util::parser::AssocOp;
	AssocOp::from_ast_binop(binop.to_hir_binop().into()).precedence()
	};
	let op_precedence = precedence(op);
	let formatted_op = op.to_hir_binop().as_str();

They're not as trivially-removable, but they're both in rustc_middle so I made it pub(crate) to at least discourage more uses.

[rustbot]

Some changes occurred in compiler/rustc_codegen_gcc

cc @antoyo, @GuillaumeGomez

[compiler-errors]

r? compiler-errors @bors r+ rollup

[bors]

📌 Commit 8ee3d29 has been approved by compiler-errors

It is now in the queue for this repository.