Quick fix Lazy nn.Graph input/output OpConf.BlobConf.is_dynamic

[chengtbf]

单卡模式下，Lazy 也是 Consistent Graph，所以这时候把 input、output 设置 is_dynamic 标记为 true 会使得系统中一些检查过不去。这个问题的本质原因是 is_dynamic 这个标记是过时的，kernel 不需要通过 blob 的标记来判断自己要不要 infer shape，而是可以通过 parallel num > 1 来决定。未来会移除 is_dynamic 标记。此处先 quick fix 一下让 nn.Graph 构出的 job 的 input output is_dynamic 标记为 False。

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.2ms (= 7007.9ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.7ms (= 6387.3ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 140.2ms / 127.7ms)

PyTorch resnet50 time: 83.8ms (= 4189.8ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.2ms (= 3708.3ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 83.8ms / 74.2ms)

PyTorch resnet50 time: 57.3ms (= 2865.7ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 48.5ms (= 2426.4ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.18 (= 57.3ms / 48.5ms)

PyTorch resnet50 time: 48.1ms (= 2406.7ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 38.5ms (= 1923.3ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.25 (= 48.1ms / 38.5ms)

PyTorch resnet50 time: 45.9ms (= 2296.2ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.9ms (= 2395.0ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 0.96 (= 45.9ms / 47.9ms)