[Speed]"merge softmax kernel"

[dzhwinter]

A future more work on #8594
In our current implementation, a lod describe the Tensor in several blocks, then every range in the lod will call Cuda kernel once.
However, this implement doesn't optimize enough, because Cuda kernel launch time is far more than the Cuda kernel execution time. So I merge these operations into one Cuda kernel to accelerate the sequence_softmax kernel.

[paddle-bot-old]

Since you haven't replied for a long time, we have closed this issue/pr.
If the problem is not solved or there is a follow-up one, please reopen it at any time and we will continue to follow up.
由于您长期未回复，我们将关闭这个issue/pr。
若问题未解决或有后续问题，请随时重新打开，我们会继续跟进。