make fused_multi_transformer support dynamically set the cache_kvs' shape and support input prefix_caches. (#46777)

* make fused_multi_transformer support dynamically set the cache_kvs' shape and support input prefix_caches.

* add LogNormal API

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* add comment

* fix bug

* fix docs

* fix bug

* fix bug

* fix bug

* add test

* add test

* change the args type of Normal sample

* fix bug

* fix bug

* fix bug

* fix bug

* add test

* add test

* format

* add comment

* add comment

* add comment

* add comment

* format code

* fix bug

* fix bug

* fix bug

* add comment

* remove name parameter for LogNormal

* organize imports

* [docs] add ipustrategy Hyperlink

* fix ipu_shard_guard docs; test=document_fix

* [docs] add set_ipu_shard note

* [docs] fix hyperlink

* update framework.py

* fix mlu_places docs; test=document_fix

* fix put_along_axis docs; test=document_fix

* fix flake8 W293 error, test=document_fix

* fix typo in typing, test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>
Co-authored-by: NNyakku Shigure <sigure.qaq@gmail.com>

* op migrated, Copy(OneDNNContext, ...) added

* mutable_data & op registration in fluid removed

* refactoring

* OneDNNGetDataType to uppercase

* missing cpu check added, handler moved to .h file

* name changed to transpose_grad

* Copy changed back to TensorCopy

* Resizing corrected, Copy(OneDNNContext) removed

* fix:gather op

* add ut

support fp16 for deformable conv

* enable hard_swish_grad unit test

* remove unused argument

* update comp cost and completion for gpt auto search

* add unittest

* fix some doc bug test=document_fix

* fix some docs issues, test=document_fix

* beta -> \beta in softplus

* threshold -> \varepsilon in softplus

* parameter name

* delta -> \delta in smooth_l1_loss

* fix some docs test=document_fix

* fix docs test=document_fix

* fix docs && 增加空行 test=document_fix

* Update python/paddle/nn/functional/activation.py, test=document_fix

* Update python/paddle/nn/layer/activation.py, test=document_fix
Co-authored-by: NSigureMo <sigure.qaq@gmail.com>

* add lookup_table op_convert, add lookup_table plugin

* support pure bfloat16

* support bf16 linear

* update PR to pass CI

* tiny fix where_grad_kernel.cu

* add bfloat16 to selu_grad to pass CI

* fix selu grad compilation error

* Move valid check from python to kernel

* fix error throw

* fix

* invalid label check

* fix

* Revert "fix"

This reverts commit 79fad6799cfa4b30423dbc84e67d7d843d22b84a.

* Revert "invalid label check"

This reverts commit 402a9707390ad5386b3222e85844b92d2e9b9fa4.

* Revert "fix"

This reverts commit 09ba3080ee0587447f875c19cdf060485f15ae3b.

* Revert "fix error throw"

This reverts commit a901bfcc2179d5c120ec29af766f392b122dab52.

* Revert "Move valid check from python to kernel"

This reverts commit baa03cc4ef82d8d45516c30dfb52bf5aead30748.

* final fix

* fix

* fix

* 之前的pr合并了大量错误代码，重新提交一份

* 之前的pr合并了大量错误代码，重新提交一份

* 修正格式问题

* 改回原来的格式

* 按照要求修改

* 按照要求修改格式

* 修复注释的问题

* 更新格式

* 测试自动格式化

* 修正英文注释

* fix docs build error

* pre-commit

* for docs build

* for docs build

* 修复mantissa计算错误的bug

* 修复误判exponent可能存在负数，导致计算量增加的情况
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

Add index_select, index_select_grad, reduce_min kernel and their unittests for kunlun. Add registers of index_select, index_select_grad, reduce_min, sqrt, sqrt_grad to xpu2_op_list.test=kunlun. (#46557)

* [CodeStyle][F401] remove unused import in unittests/collective

* empty commit, test=document_fix

* empty commit

* remove calibration file path

* remove useless code

* [AutoParallel] fix reshard when train with eval

* fix mppp