[Kunlun]Multi xpu dygraph performance optimization , add distributed.spawn support for multi xpu and some bug-fixes (#31130)

* update, test=develop

* fix fleetrun heter ps on paddlecloud

* test=develop, fix geo sgd communicator

* test=develop, gloo_init_method

* test=develop, bug fix for gloo http_init

* test=develop, bug fix for parameter_recv
* test=develop, for unittest, test_fleet_rolemaker_new

* refine fleetrun.ps_launch

* update fleet run for multi device support

* ps_graph support ps-gpu

* fix heter save

* add heter save unittest

* fix unittest & simple code

* update fleetrun

* fix fleetrun

* fix launch barrier

* fix role maker

* add paddlecloud rolemaker unittest

* rename heter_worker_device_guard

* fix worker endpoints

* fix gloo wrapper for hdfs

* GPU fleetrun support gloo

* parameterserver fleetrun support gloo

* fix get server endpoint

* fix ports conflict when launching multi-nodes in paddlecloud;test=develop

* add DISTRIBUTED_TRAINER_ENDPOINTS env for cloud;test=develop

* refine launch and distributed repr string for print

* print detailed and clear log infos; test=develop

* move paddle.fleet to paddle.distributed.fleet

* support multi-ps training mode for fleetrun; test=develop

* add fleetrun command for distributed running; test=develop

test=develop

* fix the compatibility of PY2 and PY3 in paddle.distributed.launch
test=develop

* only pull log of local rank 0
test=develop

* log exception if UnicodeEncodeError occurs when pulling log in
paddle.distributed.launch
test=develop
Co-authored-by: NSunGaofeng <peakbee@gmail.com>

test=develop

replace join to terminate

* add DataLoader, Dataset, BatchSampler