Fix masked select broadcast bug

[BBuf]

eager下masked_select不支持某些广播的问题。

close #7983

[BBuf]

这个check是不正确的，在下方提供的例子中mask的shape是(3, 3)，但这种情况也可以广播，并且广播的check逻辑在mul op里面有了，所以这个check是错误并且不必要的。

[github-actions]

View latest API docs preview at: https://staging.oneflow.info/docs/Oneflow-Inc/oneflow/pr/7984/

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

✔️ OneFlow resnet50 time: 128.8ms (= 12882.9ms / 100, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 141.2ms (= 14120.7ms / 100, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.10 (= 141.2ms / 128.8ms)

OneFlow resnet50 time: 79.4ms (= 7942.9ms / 100, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 83.3ms (= 8327.5ms / 100, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.05 (= 83.3ms / 79.4ms)

OneFlow resnet50 time: 52.7ms (= 10537.4ms / 200, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 62.0ms (= 12392.0ms / 200, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.18 (= 62.0ms / 52.7ms)

OneFlow resnet50 time: 42.7ms (= 8546.7ms / 200, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 45.7ms (= 9142.3ms / 200, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.07 (= 45.7ms / 42.7ms)

OneFlow resnet50 time: 35.4ms (= 7071.9ms / 200, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 39.6ms (= 7915.8ms / 200, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 1.12 (= 39.6ms / 35.4ms)

OneFlow swin dataloader time: 0.270s (= 54.095s / 200, num_workers=1)
PyTorch swin dataloader time: 0.249s (= 49.755s / 200, num_workers=1)
✔️ Relative speed: 0.920 (= 0.249s / 0.270s)

OneFlow swin dataloader time: 0.067s (= 13.476s / 200, num_workers=4)
PyTorch swin dataloader time: 0.067s (= 13.489s / 200, num_workers=4)
✔️ Relative speed: 1.001 (= 0.067s / 0.067s)

OneFlow swin dataloader time: 0.037s (= 7.411s / 200, num_workers=8)
PyTorch swin dataloader time: 0.036s (= 7.281s / 200, num_workers=8)
✔️ Relative speed: 0.983 (= 0.036s / 0.037s)

✔️ OneFlow resnet50 time: 136.0ms (= 13597.7ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 159.7ms (= 15974.5ms / 100, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.17 (= 159.7ms / 136.0ms)

OneFlow resnet50 time: 87.2ms (= 8717.2ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 98.9ms (= 9890.9ms / 100, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.13 (= 98.9ms / 87.2ms)

OneFlow resnet50 time: 60.4ms (= 12082.2ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 75.3ms (= 15062.3ms / 200, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.25 (= 75.3ms / 60.4ms)

OneFlow resnet50 time: 53.1ms (= 10617.4ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 67.2ms (= 13438.0ms / 200, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.27 (= 67.2ms / 53.1ms)

OneFlow resnet50 time: 49.9ms (= 9981.7ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 68.2ms (= 13641.6ms / 200, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.37 (= 68.2ms / 49.9ms)