skip sync when abnormally exiting

[daquexian]

在非正常退出时跳过退出时的多进程同步，避免出现次生 bug （例如另一个进程正常执行到了一个需要多进程同步的操作，就会产生死锁，并且会超时错误）

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.7ms (= 7033.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.8ms (= 6391.3ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 140.7ms / 127.8ms)

PyTorch resnet50 time: 83.0ms (= 4151.2ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.4ms (= 3719.2ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.12 (= 83.0ms / 74.4ms)

PyTorch resnet50 time: 55.0ms (= 2751.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.3ms (= 2363.6ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.16 (= 55.0ms / 47.3ms)

PyTorch resnet50 time: 46.5ms (= 2323.1ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 42.6ms (= 2129.6ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 46.5ms / 42.6ms)

PyTorch resnet50 time: 44.5ms (= 2227.1ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 38.3ms (= 1916.5ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.16 (= 44.5ms / 38.3ms)