Fix naive b2p error

[clackhan]

修复local rank和copy functor误用问题

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 139.6ms (= 6981.6ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.6ms (= 6377.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 139.6ms / 127.6ms)

PyTorch resnet50 time: 83.5ms (= 4176.4ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.3ms (= 3714.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.12 (= 83.5ms / 74.3ms)

PyTorch resnet50 time: 60.1ms (= 3004.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 49.7ms (= 2486.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.21 (= 60.1ms / 49.7ms)

PyTorch resnet50 time: 49.4ms (= 2467.5ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 44.3ms (= 2217.0ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.11 (= 49.4ms / 44.3ms)

PyTorch resnet50 time: 44.3ms (= 2217.1ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 41.1ms (= 2055.2ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.08 (= 44.3ms / 41.1ms)

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 139.8ms (= 6991.0ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.8ms (= 6390.9ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 139.8ms / 127.8ms)

PyTorch resnet50 time: 86.0ms (= 4299.3ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.1ms (= 3706.6ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.16 (= 86.0ms / 74.1ms)

PyTorch resnet50 time: 65.8ms (= 3288.4ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.9ms (= 2392.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.37 (= 65.8ms / 47.9ms)

PyTorch resnet50 time: 47.4ms (= 2370.8ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 44.0ms (= 2202.1ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.08 (= 47.4ms / 44.0ms)

PyTorch resnet50 time: 43.4ms (= 2170.9ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 37.1ms (= 1853.1ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.17 (= 43.4ms / 37.1ms)