modify the implementation of save_persistables and save_inference_model for fleet collective mode (#20802)

* modify the implementation of  save_persistables and save_inference_model functions for fleet collective, test=develop

* add ut, test=develop

test=develop

* fix PaddleCloud Role maker & add warning in distribute transpiler  & change rpc_retry_times

* support dump param to afs
test=develop

* code style
test=develop

* code style
test=develop

* dump param
test=develop

* dump param
test=develop

* dump param
test=develop

* dump param
test=develop

* no longer need to define all embedding layers (no one less) of all slots in each program. make trainer_param repeated in ps.proto.
* add find_distributed_lookup_table_grads instead of hard code GRAD
* support embedding stop gradient. push sparse has error before fix this.* 
* fix fill sparse, skip slots which do not have embedding. each slot's embedding in a sparse table should be used in all training programs before fix this.
* fix pull sparse, skip slots which do not have embedding.
* fix collect feasign label info, skip slots which do not have embedding.
* support when there are multi sparse tables in one or multi training programs, each program can pull/push its own related sparse tables instead of all sparse tables.
* test=develop

* add check nan / inf in downpour worker during training
* test=develop

* test=develop,Fix communicator slow bug

* test=develop, delete if() in stop_worker()

* test=develop

* fix UT, test=develop

* fix bug in fetch handler, test=develop

* fix bug in fetch handler, test=develop

* test=develop, fix fetch barrier bug

* test=develop, bug fix

* test=develop, bug fix

* test=develop, fix bug

* special case: strategy is None

* support dump multi file
test=develop

* dump fix num file
test=develop

* fix fc sort . test=develop

* refector geo sgd & communicator

* support change shuffle thread num
* support change train thread num
* fix receive shuffle data of each channel
* data norm stop gradient
* add check thread_tensor type and root_tensor type when merge metric
* remove sleep in shuffle, add config
* add config of pslib client to client communication
* fix xbox str
* add data norm op testcase
* add flush in trainer finalize

* add recompute based checkpoints methods for large batch training
test=develop

* add append_backward_with_forward_recomputation
test=develop

* refine optimizer
test=develop

* update backward and optimizer
test=develop

* make Variable usable
test=develop

* add recompute code

* refine optimizer
test=develop

* refine addup _append_backward_ops_with_checkpoints_
1) for recompute part, just cache the grad_op_desc without appending to block
2) before appending grad_op_desc to backward part, addup_repetitive_vars, remove unused branch
test=develop

* make method private

* add recompute strategy into DistributedStrategy
test=develop

* checkpoint version3
test=develop

* remove some print information
test=develop

* remove unused sumop
test=develop

* try to fix recompute with graph building modules

* add input names to vars should be held

* add memory debug tool

* backup backward

* Fix bugs

* add backward desc for op not in any segments

* add exception info for sub_block

test=develop

* modify code style

test=develop

* modify code style

test=develop

* remove print functions

test=develop

* add API spec

test=develop
test=document_preview

* make Recompute a child class of Optimizer

test=develop
test=document_preview

* add API spec

test=develop
test=document_preview

* modify API spec

test=develop
test=document_preview

* add document for Recompute

test=develop
test=document_preview

* change API doc of Rcompute

test=develop
test=document_preview

* code cleaning

test=develop
test=document_preview

* modify API spec

* fix bugs when segments hold no element

* add testcase for Recompute Optimizer

test=develop
test=document_preview

* add test for apply_gradient, and code cleaning

test=develop
test=document_preview

* add test case for load function

* enable CI

test=develop
test=document

* add test case

test=develop
test=document_preview

* add sample code for 4 function of recompute optimizer

test=develop
test=document_preview

* optimize cloud rolemaker, test=develop

change _origin_program test=develop

* support preload thread
* sleep before fleet wrapper exit for pslib core dump
* optimize hdfs log
* fix master+patch bug

Fix float16 optimizer

* fleet api add input check, test=develop

* test=develop,  communicator merge add => merge average

* add thread scope stat accurate metrics test=develop

* fix style

* fix style

* fix style

* fix style test=develop

* fix style test=develop

* fix style test=develop

* fix style test=develop

* fix style test=develop

* fix style test=develop

* fix style test=develop

* fix conflict

* fix style

* fix style test=develop

* fix error test=develop

* fix error test=develop

* dump slot

* test

* proto

* dump slot

* test

* proto

* code style

* code style

* code style

* style

* add delete after unseen days

* add unseen days

* code style

* conflict solve
test=develop

* add clear model

* code style
test=develop

* code style
test=develop

* support debug tensor of each ins
test=develop

* support debug tensor of each ins
test=develop

* learning rate

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style

* code style
test=develop

* code style
test=develop

* unitest

* style

* style

* multi phase

* add channel

* code style

* style

* style

* unitest

* style

* define

* define
test=develop

* style
test=develop

* rm define
test=develop

* linux

* linux
test=develop

* style
test=develop

* output format
test=develop

* windows ci
test=develop

fleet_desc sort fc name by dictionary sort, but we want to sort by number.

test=develop

* fix correctness of the communicator

* fix a bug in send thread when sending var context is empty, test=develop

* add lookup_table_prefetch_op and prefetch optimize, test=develop

* remove remote prefetch GPU supported

* word2vec force with CPU, test=develop

* test dist remote lookup table force with CPU, test=develop

fix fleet_desc dense_table unsort bug ，not  support format for abacus hotstart yet.

 add fleet_desc config feature & multi_sparse table,

node_num is not needed for users, so remove them and fix the bugs about it!

* fix default value in ps_pb2.py:   delta_keep_days 30 -> 16
* test=develop

* add get_last_save_xbox_base/get_last_save_xbox
* fix fleet_util bug of load paddle model
* add doc string in fleet api

* fix default value of fleet desc, default values are same with jingpai
* print log when save model

Polish fleet API to support cuda collective mode and nccl2 mode

add save cache model api in fleet& add slots shuffle in dataset module & add metric op to calculate ctr related metrics (#18871)

* add ctr related metric layer test=develop

* add save cache and slots shuffle test=develop

* add save cache and slots shuffle test=develop

* fix error

* fix error

* fix style for ci

* fix for comments

* change SlotsShuffle input to std::strinf for generality

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix stylr

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* fix style

* change non-const reference to pointer

* fix style

* fix style

* fix style test=develop

* fix style  test=develop

* add return ins num in ctr metric op

* change dtype to float in metric_op.py

* fix error test=develop

* fix style test=develop

* fix API spec

* fix API spec

* fix API spec test=develop

* add UT test=develop

* add fleet util (fleet/utils/fleet_util.py): functions for users' convenience
* add some interface in hdfs util : hdfs is_file、hdfs cat

* support filelist size < trainer num
* pull dense when stop, to make sure local dense params are same as pserver, so save paddle model will save dense model same as pserver
*  enable QueueDataset train same filelist for serveral times

adjust ins weight according to nid slot , user can specify adjust_ins_weight in strategy

(1) set fleet_send_batch_num a default value according to trainer num， the previous 80000 is fixed，if trainer num is much less or larger than 100，global shuffle may have timeout error.

(2) fix load one table bug, add barrier