Fix bug error msg format

[liufengwei0103]

1. 重写了语义更明显的函数名
2. 重写了把msg缩短的函数
3. 加上了在error_type中的msg信息

[daquexian]

可以看下这个可不可以用反射实现，这样手动列出每一种情况是不可维护的，用反射的话伪代码大概是这样：

ss << PbMessage2Txt(error->getMessage(error->error_type_name()))

[liufengwei0103]

可以看下这个可不可以用反射实现，这样手动列出每一种情况是不可维护的，用反射的话伪代码大概是这样：

ss << PbMessage2Txt(error->getMessage(error->error_type_name()))

已修改采用反射实现

[MARD1NO]

这里是否需要包一层CHECK？

[MARD1NO]

这里是否也要check下

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 139.6ms (= 6980.8ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.1ms (= 6405.5ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 139.6ms / 128.1ms)

PyTorch resnet50 time: 84.9ms (= 4247.4ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.5ms (= 3724.3ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.14 (= 84.9ms / 74.5ms)

PyTorch resnet50 time: 56.4ms (= 2819.8ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.8ms (= 2391.5ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.18 (= 56.4ms / 47.8ms)

PyTorch resnet50 time: 49.5ms (= 2477.5ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 40.3ms (= 2016.0ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.23 (= 49.5ms / 40.3ms)

PyTorch resnet50 time: 42.5ms (= 2125.0ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 43.2ms (= 2158.5ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 0.98 (= 42.5ms / 43.2ms)

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 141.5ms (= 7076.9ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.2ms (= 6408.0ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.10 (= 141.5ms / 128.2ms)

PyTorch resnet50 time: 84.3ms (= 4212.5ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.5ms (= 3722.9ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.13 (= 84.3ms / 74.5ms)

PyTorch resnet50 time: 54.9ms (= 2744.4ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 48.2ms (= 2408.9ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.14 (= 54.9ms / 48.2ms)

PyTorch resnet50 time: 48.6ms (= 2429.9ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 39.8ms (= 1988.5ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.22 (= 48.6ms / 39.8ms)

PyTorch resnet50 time: 45.4ms (= 2270.3ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 36.4ms (= 1818.9ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.25 (= 45.4ms / 36.4ms)