Log_softmax backward case#1: axis=-1

[AshburnLee]

PR types

Performance optimization

PR changes

OPs

Describe

实现log_softmax的cuda版本。如下是前向计算的3个case。当前PR实现case#1 的反向计算。

if (inner_size == 1) {
    if (dim_size <= 1024 && dim_size * sizeof(T) <= 4096) {
        case#1  
    } else {
        case#2
    }
} else {
    case#3
}

[paddle-bot-old]

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[Xreki]

LGTM. 有3处小的修改建议，可以下个PR再改。