* set input type to fp16

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

* add isort config

* isort all files

* update config

* re-blacken python code

* temporarily disable date and diff_py_file

* skip a format

* Add missing fp32 config and reduce the testing combination

* Reduce trt matmul pass test max examples

* [CodeStyle][F401] remove unused imports in unittests/dygraph_to_static

* [CodeStyle][F401] remove unused imports in unittests/ir

* add noqa after required imports

* Reduce pool2d test configuration

* Reduce depthwise_conv2d test configuration

* Reduce trt_convert_conv2d_fusion test configuration

* Reduce trt_convert_conv2d test configuration

* Reduce trt_convert_conv2d_transpose test configuration

* Reduce trt_convert_hard_swish test configuration

* Enhance trt auto scan test error message and mechanism

* Increase FP16 trt ut tolerance

* use yapf to format all python file

* yapf exclude two unittests file for they rely on writing and reading file, and format will break them

* disable diff_py_file because too many diff files cause command following failed

* add flatten/flatten2 converter test cases

* add fatten/flatten2 trt converter test cases