* Support memory eager deletion on recurrent OP (#17710)

Test PaddingRNN on V100 GPU device.

Test configuration: large model, padding mode (which is the mode using recurrentOp), one GPU.
                   
GPU memory (MiB):   6414 (this PR)     vs   6837 (without this PR)
Speed (steps/s):         10.28 (this PR)    vs    9.89 (without this PR)

* Fix random test_recurrent_op failure (#18718)

The change includes 3 things:

1. Set CPU_NUM to 1 in the tests because the ParallelExecutor will print warning that CPU_NUM is not set and use default 1.

2. Old tests compare two RNNs, hand written simple RNN and same RNN built by Paddle, but initialized RNN weights in numpy random and Paddle random separately. Fixed it by setting weights and bias values.

3. Also set numpy random seed in the tests. Now the two RNNs diff can be smaller (rtol from 0.1, 0.2 to. 0.01) in the tests.

* Revert "Revert "Fix allocator bug""

This reverts commit 174d0d0b.

* Revert "fix travis ci"

This reverts commit 5656fa9f.

test=develop

* add inlined_vector.h, test=develop

* add inlined_vector_test,test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

support win32 flag in io.cc shell.cc, fix code style problem in fleet_wrapper, fix lodtensor_printer_test problem
test=develop

test=develop

add include headers to fix travis-ci
test=develop

modify allocator strategy
remove changes of legacy buddy_allocator
test=develop

test=develop

test=develop

* wip multi process multi gpu dist training

* workable for p2p

* update test=develop

* change back env name test=develop

* fix alloc init

* fix cpu build test=devlop

* fix mac tests test=develop

* refine code

* refine test=develop

* human readable memory warns test=develop

* update test=develop

* refine test=develop

* fix build test=develop

- add naive executor
- fix concurrency performance issue

* Fix cpplint errors in paddle/fluid/string

* Fix unit test error

* Correct