fix weight decay in RMSprop

[L1aoXingyu]

fix unittest rmsprop numpy implementation without weight decay
passing weight decay as l2 in rmsprop optimizer

[github-actions]

CI failed, removing label automerge

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 141.2ms (= 7061.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 127.7ms (= 6386.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.11 (= 141.2ms / 127.7ms)

PyTorch resnet50 time: 83.3ms (= 4164.7ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.2ms (= 3708.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.12 (= 83.3ms / 74.2ms)

PyTorch resnet50 time: 56.5ms (= 2824.9ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.6ms (= 2381.2ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.19 (= 56.5ms / 47.6ms)

PyTorch resnet50 time: 48.5ms (= 2423.8ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 40.5ms (= 2025.4ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.20 (= 48.5ms / 40.5ms)

PyTorch resnet50 time: 39.8ms (= 1989.3ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 41.5ms (= 2072.8ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 0.96 (= 39.8ms / 41.5ms)