* Support Mixed Precision training in @to_static

* fix block.vars logic

* fix GPU training loss diff

* remove unused code

* Add build_strategy in @to_static to support open pass

* fix os.environ

* add timeout

* disable test_build_strategy on openblas

Revert "Revert "[Dy2Stat] Refactor ExecutorCache logic and pre-support BuildStrategy for pass (#34181)" (#34348)" (#34384)

This reverts commit 577fdde5.

Revert "[Dy2Stat] Refactor ExecutorCache logic and pre-support BuildStrategy for pass (#34181)" (#34348)

This reverts commit 609f8225.

* modify into program_id

* fix cache_info declare problem

* fix python int to C long problem

* modify point to reference

* add ENVS

* support Nest sequtial container

* rename model path

* cache core.ops, test=develop

* refine, test=develop

* jit.save/load support method with parameters.

* add unittest and warning

* polish warning message.

* fix paddle.no_grad

* fix paddle.no_grad

* reduce sum op default fp32, add into amp black list

* reduce_sum default fp32 can avoid return inf when the sum value large than 65504

* add get and set for Grad_scaler

* refine some API name and comments

* refine API name and comments

* refine some comments

* refine temp_scope_vec logic

* polish partial_program

* fix fake var

* add stop_gradient in spec

* fix fake_var

* fix unittest

* support type hint

* fix unittest

* transform complex scale to tensor

* add test_case for complex scalar

* modify import paddle

* Combine amp and qat
* add unit test

* del py2 code2

* fix test timeout

* support non-tensor type

* fix unittest failed

* add unittest with prune

* rm unused code

* coverage

* fix two or

* Support append customize attributes into op_desc in nn.Layer

* fix code style

* support override

* add unittest

* Modify into core.ops.run_program

* add DDout in core.ops.run_program

* fix typo

* add DOut

* fix typo

* put DOut last

* modify API nn.Bilinear's doc, test=develop

* move data to CUDAPlace in advance

This PR made these changes to support double grad:

1. Translate `paddle.grad` to `paddle.static.gradients` to support double grad for dy2stat.
2. Fix IfElseTransformer bug which may not change value if "Store before Load" variable is in "Store" statement is in IfElse conditional statement
3. Add `DOut` to support double grad variables in `run_program_op`
4. Add support for renaming for double grads for `jit.save/load`

* fix the bug of div operation result using scale method do not exactly equal the result of elementwise_div method

* remove __div__ , __rdiv__ methods which do not define in python3

* modify the note

* add test case

* add test case

* Support convert sublayers in Sequential Container

* remove paddle.jit.set_code_level

* BugFix StaticAanlysis with gast.Subscript

* remove codes

* Add param_guard in ParameterList to support @to_static

* Refactor param_guard of @to_static

* fix unittest failed

* add more unittest

* add varbasecopy func to fix the paraBase type bug in layers.to API

* overload _copy_to func for ParamBase

* add xpuplace

* add waiting varbsecopy completion when not blocking

* fix dst_device bug

* modify varbase to shared_ptr

* fix find_unused_parameters default value

* Add deprecated warning info.

* Add unittest for deprecated decorator.

* Add warning info for tensor.grad

Dy2stat failed when user writes return paddle.to_tensor(xxx), the reason is that visit_Expr doesn't work when the Expr is in return. Some other statements may trigger same bug. To fix it, we re-wrote a transformer to transform paddle.to_tensor to paddle.assign for all Call nodes.