update clang tidy diff warnings-as-errors option

[daquexian]

-warnings-as-errors 跳过 maybe-glog-fatal 检查、在 ci 中开启 warnings-as-errors

[PragmaTwice]

LGTM

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 141.2ms (= 7057.7ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.4ms (= 6420.3ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 141.2ms / 128.4ms)

PyTorch resnet50 time: 85.0ms (= 4248.4ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.9ms (= 3747.2ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 85.0ms / 74.9ms)

PyTorch resnet50 time: 58.7ms (= 2934.9ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.7ms (= 2383.2ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.23 (= 58.7ms / 47.7ms)

PyTorch resnet50 time: 47.7ms (= 2382.8ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 39.5ms (= 1974.3ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.21 (= 47.7ms / 39.5ms)

PyTorch resnet50 time: 44.4ms (= 2220.8ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 36.4ms (= 1821.8ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.22 (= 44.4ms / 36.4ms)