Fix LayerNorm expr bug

[wyg1997]

顺手修复 #6106 中 LayerNorm 打印的 bug

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 128.3ms (= 6416.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 140.7ms (= 7035.0ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.10 (= 140.7ms / 128.3ms)

OneFlow resnet50 time: 74.7ms (= 3734.9ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 84.7ms (= 4236.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.13 (= 84.7ms / 74.7ms)

OneFlow resnet50 time: 49.9ms (= 2497.0ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 58.9ms (= 2944.5ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.18 (= 58.9ms / 49.9ms)

OneFlow resnet50 time: 39.7ms (= 1985.6ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 49.2ms (= 2461.8ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.24 (= 49.2ms / 39.7ms)

OneFlow resnet50 time: 42.1ms (= 2106.4ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 40.1ms (= 2006.1ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 0.95 (= 40.1ms / 42.1ms)

OneFlow resnet50 time: 138.8ms (= 6938.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 148.9ms (= 7445.5ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.07 (= 148.9ms / 138.8ms)

OneFlow resnet50 time: 87.1ms (= 4357.3ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 88.3ms (= 4415.2ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.01 (= 88.3ms / 87.1ms)

OneFlow resnet50 time: 62.8ms (= 3139.8ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 67.1ms (= 3354.8ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 1.07 (= 67.1ms / 62.8ms)

OneFlow resnet50 time: 60.3ms (= 3014.2ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 57.3ms (= 2865.7ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 0.95 (= 57.3ms / 60.3ms)

OneFlow resnet50 time: 59.4ms (= 2969.3ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow GPU used (rank 0): 0 MiB
PyTorch resnet50 time: 49.5ms (= 2477.0ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
PyTorch GPU used (rank 0, estimated): 0 MiB
Relative speed: 0.83 (= 49.5ms / 59.4ms)