Fix a bug in save_inference_model and prune when the program is initailized by load_inference_model (#10011)

* Fix bug in save_inference_model and prune when the program is initialized by load_inference_program.

* Save the transpiled program instead.

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 is_test issue

* add paddle enforce

* fix bug

* add new func

* small fix

* address comments

* merge fluid dist tests

* update cmake

* Generated functions can take `*args` as inputs.

Use a new scope for inference in python unittest to avoid changing the value of variables for training.

* First basic implementation

* Add infer example for image_classification

* Address review comments: round 1

* "fix decode bug"

* "follow commnet"

* "fix error"

* "fix hook bug"

* fix based comment

* fix copyright

* fix based on comment

* add copyright hook

* add copyright hook

* refine copyright hook

* "test copyright hook"

* fix check style

* fix ci

* Make cast op support bool

Also add `elemwise_sub/mul/abs/clip` layers

* Make fuild.layers as a module

* Move layers as a module

* Split layers.py into layers module

* Fix CI

* Fix CI

* Add DataFeeder

A v2 API like data feeder for book demos.
We can feed data directly from reader.

* Fix CI

* Remove batch_size_dim for feeder

Also add __all__ to data_feeder.py

* Follow comment

* batch norm support matrix input

* update gpu code

* format code

* py_test support argument, test_image_classification_train support argument

* use REMOVE_ITEM to rm item from list in cmake

Change books just use `import fluid`, not submodules

* Stash

* Stash

* Polish Evaluator

* Merge code

* Revert

* is_training to is_test in dropout op

* handle dropout and batch_norm operator when prune pdesc in testing mode

* handle dropout and batch_norm operator when prune pdesc in testing mode

* add get_inference_program method

* fix dropout op

* fix ci

* test data after each batch training

* refine code

* refine test_book3

* fix ci

* follow comments

* Change all `data_type` in Python to `dtype`

* Change `date_type` in C++ to `dtype`

* Refine

- Removed all main_program and startup_program in the demo.
- Using paddle.default_main_program() hides the implementation detail (e.g., using g_main_program) from the user, we can change the implementation in the future much easier.

* init commit

* change some dir name

* add acc layer
* memory log level change from 3 to 10
* use gaussian random to init conv parameters
* use initializer
* fix import
* batch_norm use helper to create persistable var
* refine code
* train only 2 batches for test
* use g_program and g_init_program
* use XavierInitializer to init fc parameter

* optimizer use init_program

* create persistable variable

* add create_persistable_var to block

* optimizer use create_persistable_var

* fix prefix

* move create_global_persistable_var from Block to LayerHelper

* Polish Optimizer initialization code.

* Using the LayerHelper to create initialize operator and variables

* add_accumulator should use an independent data type

* default use param data type for accumulator

* add resnet

* optimize code