* [dtype] add fp16 support for dist_kernel

* fix typo

* fix CE

* fix CE

* fix CE

* fix CE

* fix CE

* refactor

* fix CE

* fix CE

* fix varname

* add bf16

* add ut for bf16

* fix CE

* Improve GC for pipeline parallel

* Delete print

* fix bug of nop_op and sharding

---------
Co-authored-by: Nchenruibiao <chenruibiao@baidu.com>

* support new ir load combine

* update

* polish code

* remove print

* polish code

* fix bug

* polish code

* fix compile bug

[clang-tidy] Open clang-analyzer-unix.Vfork、clang-analyzer-security.insecureAPI.vfork Check (#56253)

* [Fluid] Migrate number_count to PHI

* fix out alloc

* fix ut (add python_api)

* refine code

* refine code

* refine code

* [CustomOP] InferDtypeFn supports attrs

* Update paddle/fluid/framework/custom_operator.cc

* update cmake list

* fix cpu device

* change unittest time

* op_vjp generate template

* op_vjp_gen templete

* delete print
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

---------
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

* [AMP] add fp16 op detection for custom device

* resolve conflicts

* add code

* fix bug

* fix bug

* delete unused code

* refine code

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* add ir output check in OpTest

* add ir grad check in op test

* fix bug in output check

* trigger CI

* test gpu ci

* trigger CI

* trigger CI

* add white list to relax precision check for some tests

* relax timeout of test_concat_op

* relax timeout of test_concat_op

* add rmsnorm residual bias add and quant

* refine python interface

* add rmsnorm unittest

* Add layernorm

* fix layernorm unittest

* refine unittest

* fix example code

* fix review comment

* Add ShapeDialect dict & SymbolicDimOp without UT.

* add unittest and fix Update_xxx_Func.

* change std::string to const std::string & and remove phi dependency.

* add reshape spmd rule

* add unit test for reshape spmd rule

* bug fix

* replace the print_info function with to_string

* fix typo

* bug fix

* add handling for "0" in target shape

* remove the part of computing size in dim_trans.cc

* Add enforce for sparse_bn.

* Add enforce for sp conv.

* [dy2static]implement tensor.cuda() in static graph

* [dy2static]implement tensor.cuda() in static graph - change the patch place

* [dy2static]implement tensor.cuda() in static graph - fix code-block in comment

* [dy2static]implement tensor.cuda() in static graph - add ut for warning branch

* Fix the shape of input sin and cos for fused_rope.

* Update shape in unittest.

IR 的 builtin dialect 中加入 IndexType