Fix gpu compilation with cuda 10.1 related to #751

[chck]

#771 was not sufficient as the latest platform in my case.
I checked the compilation works with this PR.

$ make
...
impl/../utils/Tensor.cuh(362): error: type name is not allowed

impl/../utils/Tensor.cuh(362): error: expected an expression

Platform

OS

$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=18.04
DISTRIB_CODENAME=bionic
DISTRIB_DESCRIPTION="Ubuntu 18.04.2 LTS"

g++

$ g++ --version
g++ (Ubuntu 8.3.0-6ubuntu1~18.04.1) 8.3.0

cuda

$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Fri_Feb__8_19:08:17_PST_2019
Cuda compilation tools, release 10.1, V10.1.105