fix documentation (#102)

Author: xadupre

onnx_array_api/reference/ops/op_cast_like.py

@@ -17,7 +17,8 @@
 
 def _cast_like(x, y, saturate):
     if bfloat16 is None:
-        return (cast_to(x, y.dtype, saturate),)
+        to = np_dtype_to_tensor_dtype(y.dtype)
+        return (cast_to(x, to, saturate),)
     if y.dtype == bfloat16 and y.dtype.descr[0][0] == "bfloat16":
         # np.uint16 == np.uint16 is True as well as np.uint16 == bfloat16
         to = TensorProto.BFLOAT16