check compiler version in cmake

[daquexian]

No description provided.

[PragmaTwice]

LGTM

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.4ms (= 7019.3ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.9ms (= 6393.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 140.4ms / 127.9ms)

PyTorch resnet50 time: 85.2ms (= 4261.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.3ms (= 3716.8ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.15 (= 85.2ms / 74.3ms)

PyTorch resnet50 time: 59.2ms (= 2961.1ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 49.1ms (= 2455.0ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.21 (= 59.2ms / 49.1ms)

PyTorch resnet50 time: 49.0ms (= 2451.3ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 39.7ms (= 1983.3ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.24 (= 49.0ms / 39.7ms)

PyTorch resnet50 time: 43.1ms (= 2156.9ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 37.6ms (= 1880.0ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.15 (= 43.1ms / 37.6ms)