Cherrypick  en doc: reorder_lod_tensor_by_rank , map_reader, WeightNormParamAttr, io.compose  (#20463)

* modify map_reader en doc (#20229)

* modify WeightNormParamAttr en doc (#20218)

* modify reorder_lod_tensor_by_rank en doc

* modify compose en doc (#20217)

improve the docs of load_inference_model&load_params&load_persistables&chain. test=document_fix (#20408)

test=develop

test=develop

test=develop

test=develop

* add multiprocess_reader

* add multiprocess_reader to reader decorator

* support piped multi process reader

* revert v2 decorator

* add comment to multiprocess_reader

* optimize code

* use ujson to speed up json serialize/deserialize

* add assert to multiprocess_reader

* update comment of multiprocess_reader

* optimize ujson import, handle error case

* optimize import ujson

* remove ujson from requirements.txt

* add import sys to decorator.py

Move six.moves.queue to Queue

This reverts commit 4d8e8ee2, reversing
changes made to d6a9f005.

In this way we can build and test using WITH_FLUID_ONLY flag being set
to ON.

- move paddle.v2.dataset,reader to paddle.dataset,reader
- remove unused code (which depends on v2) in paddle.dataset,reader

fix pipe reader on multi passes

images reader: read the data without untarring the tarball file.
image.py: move batch function from reader to image.py