all: Convert mp_uint_t to mp_unary_op_t/mp_binary_op_t where appropriate

The unary-op/binary-op enums are already defined, and there are no
arithmetic tricks used with these types, so it makes sense to use the
correct enum type for arguments that take these values.  It also reduces
code size quite a bit for nan-boxing builds.

py/objstr.c

@@ -280,7 +280,7 @@ const byte *find_subbytes(const byte *haystack, size_t hlen, const byte *needle,
 // Note: this function is used to check if an object is a str or bytes, which
 // works because both those types use it as their binary_op method.  Revisit
 // MP_OBJ_IS_STR_OR_BYTES if this fact changes.
-mp_obj_t mp_obj_str_binary_op(mp_uint_t op, mp_obj_t lhs_in, mp_obj_t rhs_in) {
+mp_obj_t mp_obj_str_binary_op(mp_binary_op_t op, mp_obj_t lhs_in, mp_obj_t rhs_in) {
     // check for modulo
     if (op == MP_BINARY_OP_MODULO) {
         mp_obj_t *args = &rhs_in;
@@ -380,9 +380,10 @@ mp_obj_t mp_obj_str_binary_op(mp_uint_t op, mp_obj_t lhs_in, mp_obj_t rhs_in) {
         case MP_BINARY_OP_MORE:
         case MP_BINARY_OP_MORE_EQUAL:
             return mp_obj_new_bool(mp_seq_cmp_bytes(op, lhs_data, lhs_len, rhs_data, rhs_len));
-    }
 
-    return MP_OBJ_NULL; // op not supported
+        default:
+            return MP_OBJ_NULL; // op not supported
+    }
 }
 
 #if !MICROPY_PY_BUILTINS_STR_UNICODE