ReshapeOp support any sbp

[wyg1997]

fixes oneflow-inc/oneteam#1040

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.9ms (= 12887.6ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.7ms (= 14073.9ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 140.7ms / 128.9ms)

✔️ OneFlow resnet50 time: 75.1ms (= 7510.1ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 84.8ms (= 8481.6ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.13 (= 84.8ms / 75.1ms)

OneFlow resnet50 time: 49.0ms (= 9802.5ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 55.1ms (= 11023.3ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.12 (= 55.1ms / 49.0ms)

OneFlow resnet50 time: 44.3ms (= 8861.0ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 44.9ms (= 8986.7ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.01 (= 44.9ms / 44.3ms)

OneFlow resnet50 time: 36.2ms (= 7241.4ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 37.9ms (= 7585.2ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.05 (= 37.9ms / 36.2ms)

✔️ OneFlow resnet50 time: 143.2ms (= 14315.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 161.9ms (= 16187.3ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.13 (= 161.9ms / 143.2ms)

OneFlow resnet50 time: 87.9ms (= 8788.6ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 101.2ms (= 10123.3ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.15 (= 101.2ms / 87.9ms)

OneFlow resnet50 time: 63.3ms (= 12654.4ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.0ms (= 14797.9ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 74.0ms / 63.3ms)

OneFlow resnet50 time: 52.5ms (= 10509.8ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 62.0ms (= 12392.5ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.18 (= 62.0ms / 52.5ms)

OneFlow resnet50 time: 50.3ms (= 10051.8ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 59.6ms (= 11913.4ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.19 (= 59.6ms / 50.3ms)