* remove cached kernel context

* revert dataloader format change

* hot fix for dataloader thread error

* polish comment

* fix type in comment, test=document_fix

* fix DataLoader exit with SIGABRT/SIGSEGV. test=develop

* fix drop_last not work in IterableDataset. test=develop

* add persistent_workers. test=develop

* del py2 code2

* fix test timeout

* fix dataloader exit hang when join re-enter. test=develop

* fix dataloader exit error if user exit program when dataloader is still iterating. test=develop

* polish tensor pipeline. test=develop

* add dict/str/list supprot for DataLoader. test=develop

* fix dataloader collate return list mix tensor and numpy array. test=develop

Set expected place in child thread for dataloader to avoid costing cuda memory on other card (#30338)

* set expected place in child thread for dataloader

* set device id when set tensor from numpy

* revert tensor_py change

* add compile guard

* fix ci

* fix bug

* Update get_worker_info API

* Update dataloader_iter.py

* Update dataloader_iter.py

* Update dataloader_iter.py

* upgrade comment string to raw string

* fix string in

* fix string with ' '

* revert update on comments

* upgrade only necessary

* fix sample code checker

* fix comments with '''

* DataLoader support not auto collate batch. test=develop

* fix DataLoader return same format between static & dynamic in single mode. test=develop

* fix dataloader

* fix DataLoader single process mode exit SIGABRT error. test=develop

* move worker loop to top level

* move reader process loop to top level

* fix failed unittests

* add TensorDataset. test=develop

* fix test_multiprocess_dataloader unittest. test=develop

* add IterableDataset support in multiprocess DataLoader. test=develop

* add pin memory control

* fix buffered reader init problem

* fix unittest error

* add unittest for coverage

* make default_collate_fn visible. test=develop

* fix test_multiprocess_dataloader_static random fail. test=develop

* add DataLoader, Dataset, BatchSampler