* Fix TRT plugin registry without TRT lib (#25982)

* fix trt plugin registry without trt lib

* support trt4

* refine code style

* pick ea851796 from develop

* cherry-pick develop PR  #26273 && #27796

* fix unittest error

* fix unittest error

* remove const_cast
Co-authored-by: NPei Yang <peiyang@baidu.com>

This commit fixs the compiling bug regarding unique_ptr of IOptimizationProfile.

IOptimizationProfile has protected dtor and is controlled by TensorRT
internally. Application shouldn't delete the pointer of IOptimizationProfile.
See TensorRT document: https://docs.nvidia.com/deeplearning/sdk/tensorrt-api/c_api/classnvinfer1_1_1_i_builder.html#a9ac47e100454151d8206ac91d543299a
test=develop
Co-authored-by: NJeng Bai-Cheng <jeng1220@users.noreply.github.com>

* solve conflict

* fix crash when trt not found in python; update unittest model path

* add dynamic plugin support.
test=develop

* change emb eltwise layernorm to math function
test=develop

* add emb eltwise layernorm
test=develop

* can run dynamic shape ernie
test=develop

* fix ci
test=develop

* add ut for trt ernie dynamic

test=develop

* refine dynamic shape c++ interface.
test=develop

* fix comments
test=develop

* fix comments
test=develop

* change the ci trt from version 5. to 6.0

* paddle-trt dynamic shape support init

* conv+bias or conv+bn dynamic shape support
test=develop

* modity trt engine opconvert
test=develop

* fix ci error
test=develop

* add mutex for trt engine
test=develop

* add the test for copy_to_cpu
test=develop

added splitter "__" between weight name and suffix number to avoid conflicts.

* fix trt bugs when sharing params, test=develop

* add unittest for cascade_rcnn

* add tensorrt support for windows

* Fix Mask rcnn predictor
    1. refine memory optim algorithm to support the model with the block op.
    2. output diff : modify the affine channel fuse
    3. add condition_block_infer op
add interface for setting trt calib table dir
test=develop

* add the missing files.
test=develop

* 1 add trt fp16 support
test=develop

test=develop

* fluid int8 train and trt int8 predict align.
trt int8 predict init
op converter

* 2. align fluid int8 train and trt int8 inference.
enhance quant dequant fuse pass
enhance op converter, trt engine, trt engine op, trt subgraph pass.

* 3. add delete_quant_dequant_pass for trt

test=develop

* 4. add the missing file
test=develop

* 5. i modify the c++ interface, but forget to modify the pybind code
fix the IS_TRT_VERSION_GE bug, and fix elementwise op converter
test=develop

test=develop

1). add static trt load model
2). fix bug: when device_id is not 0, the trt will have a bug
test=develop

add simple static mode loading
test=develop

1. bind trt input and output to fluid tensors

test=develop

1). add static trt load model
2). fix bug: when device_id is not 0, the trt will have a bug
test=develop

add simple static mode loading
test=develop

1. bind trt input and output to fluid tensors

BUG: After continuing to input different data, the output cannot be aligned
test=develop

test=develop

fix comments

test=develop

test=develop

* Initialize the elementwise plugin.

* Implement the basic CUDA kernel of elementwise plugin.
test=develop

test=develop

* exhaustive search for cuDNN conv.
* Refine code and add unit testing.
* Fix model load in fluid/inference and unit testing in conv2d
* Follow comments.
* Fix compiling test=develop