Dev RMSprop graph conf

[L1aoXingyu]

• Add RMSprop conf for nn-Graph
• Add unittest for RMSprop conf

[CLAassistant]

All committers have signed the CLA.

[L1aoXingyu]

两点疑问:

• 通过看注释发现 RMSprop 当前并不支持 momentum > 0
• RMSprop for nn-Graph 是否不支持 scale != 1.0

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.6ms (= 7029.7ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.8ms (= 6389.7ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 140.6ms / 127.8ms)

PyTorch resnet50 time: 83.6ms (= 4182.4ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.1ms (= 3706.8ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 83.6ms / 74.1ms)

PyTorch resnet50 time: 58.7ms (= 2933.4ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 48.8ms (= 2438.9ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.20 (= 58.7ms / 48.8ms)

PyTorch resnet50 time: 47.8ms (= 2388.7ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 40.0ms (= 2000.9ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.19 (= 47.8ms / 40.0ms)

PyTorch resnet50 time: 42.0ms (= 2098.2ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 39.4ms (= 1971.6ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.06 (= 42.0ms / 39.4ms)