[XPU] Paddle add xpudnn library && conv_transpose alter kernel

[HarperCy]

PR Category

Operator Mechanism

PR Types

New features

Description

Paddle add xpudnn library, which is already included in xhpc output but never added to paddle compiling logic.
This library, containing many high performance convolution operators, is aimed to replace the outdated conv kernels used in current Paddle XPU Ops, should, in theory, both boost the speed and extend the generalization ability.
In this commit, I replace the original conv_transpose kernel to xpudnn ones.

[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.

[CLAassistant]

All committers have signed the CLA.

[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.

---

root 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.}

[paddle-ci-bot]

Sorry to inform you that 27f7889's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually.

[dynamicheart]

LGTM

[skywalker2012]

LGTM