[Inference] add config.enable_low_precision_io api and remove rely on AnalysisConfig::Precison in trt (#52485)

* fix transfer_layout when input size if too big
* do not add TransferLayoutKernelGPU
* add int64 and add check

* Add GPU kernel for multiclass_nms3 op

* Make multiclass_nms3 gpu kernel output consistent with cpu kernel

* Fix API incompatibility

* Fix unittests on builds without CUDA

* Fix ROCM build

* Remove fluid headers; Use default atol for unittest

* Change function and variable naming

* Add comments; Reduce redundant code

* Use paddle test framework

Print python trace back when debugmode = CHECK_NAN_INF_AND_ABORT  and backward has nan/inf  (#52808)

* bind xpu op: 3D grid sample

* fix edge cases in xpu op: reshape & slice

* add types and attributes

* remove some const_cast

* refine code

* Improve Readability and Overall Clarity of Logging

* Adds the set_input_type API for specifying input data types

* Specifying input data types

* add minimum grad composite rules

* add public python api

* fix format

* fix format

* update testcase

* fix testcase

* fix format

* fix cmakelist.txt

* fix format

* fix param problem

* fix op and composite rule

* fix bf16 cpu support problem

* fix bf16 cpu issue

* fix axis error log

* add axis for maximum

* revert commit

* remove .orig

* fix generic problem

* revert max op

* fix axis error

* fix maximum axis

* fix test_check_output

* fix cinn

* fix minimum maximum axis check

* Reorganize the forward codes of flash-attention.

* Fix forward.

* Remove some noused codes.

* Simplify codes and fix backward.

* Change all LOG(INFO) to VLOG and fix the backward.

* add scale for AF2 flash_attn, much thanks to xreki and shaojie for debug these codes

* decrease the effect of debug print on performance

* Unify the initialize of flashattn arguments.

* Rewirte the reshape of temp_mask and temp_bias.

* API support use_flash_attn.

* Fix compiling error on CI.

* Try to crop the flash-attention lib.

* Correct the condition of whether can use flash-attn.

* Remove the softmax_out argument.

* Remove is_causal.

* Polish codes.

* Fix qkv_transpose_out's shape and scaling of Q * K.

* Update commit of flash-attention.

---------
Co-authored-by: NLiu Yiqun <liuyiqun01@baidu.com>

* [tools] add PADDLE_API check file diff approvals

* [tools] fix determine

* [tools] fix determine

* [tools] Change to full character matching
Co-authored-by: N张春乔 <83450930+Liyulingyue@users.noreply.github.com>

* [tools] Update echo_line
Co-authored-by: N张春乔 <83450930+Liyulingyue@users.noreply.github.com>

* [tools] Update check_approval
Co-authored-by: N张春乔 <83450930+Liyulingyue@users.noreply.github.com>

---------
Co-authored-by: N张春乔 <83450930+Liyulingyue@users.noreply.github.com>

* remove user define grad

* fix errors

* remove unused self.x_grad, self.out_grad

* Add large dim test of log_softmax

* fix

* review

* modify opcompat bug

* modify pybind

* delete bf16 of cross entropy

* delete bf16 of cross entropy

* decrease_peak_memory