[Paddle Inference] Add multi_query_attention kernel

[superxf]

PR types

others

PR changes

OPs

Description

• Support masked multiquery attention for transformer decoder stage.
• Support dtype：[float | float16 | bfloat16]
• export python api： from paddle.incubate.nn.functional import masked_multiquery_attention

Pcard-74485

[paddle-bot]

你的PR提交成功，感谢你对开源项目的贡献!
请关注后续CI自动化测试结果，详情请参考Paddle-CI手册。
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[vivienfanghuagood]

例子建议给一个group=1和group=2的例子吧

[superxf]

Done

[luotao1]

@superxf 如果是内部同学，可以在 http://rdmp.dltp.baidu-int.com/data-management/github-user 登记下

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
1 out of 2 committers have signed the CLA.

✅ superxf
❌ lizhenyun

---

lizhenyun seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[xiaoxiaohehe001]

可以参考 #56411 ，把 .h 文件消除掉

[paddle-ci-bot]

Sorry to inform you that 22747b5's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

[carryyu]

谢谢，已收到！