Add CUDA_NVCC_THREADS_NUMBER

[liujuncheng]

No description provided.

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.0ms (= 6997.6ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.1ms (= 6406.6ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 140.0ms / 128.1ms)

PyTorch resnet50 time: 87.9ms (= 4396.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.6ms (= 3730.6ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.18 (= 87.9ms / 74.6ms)

PyTorch resnet50 time: 61.5ms (= 3073.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 50.3ms (= 2514.8ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.22 (= 61.5ms / 50.3ms)

PyTorch resnet50 time: 48.0ms (= 2401.8ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 40.3ms (= 2017.0ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.19 (= 48.0ms / 40.3ms)

PyTorch resnet50 time: 44.2ms (= 2208.9ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 36.3ms (= 1817.0ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.22 (= 44.2ms / 36.3ms)