GitOrigin-RevId: 94e333ec80

GitOrigin-RevId: adede7cef6

GitOrigin-RevId: 5c1f7c669f

GitOrigin-RevId: b383beb886

GitOrigin-RevId: 05618e5ac5

GitOrigin-RevId: 9acab0a49f

GitOrigin-RevId: 14e1de9952

GitOrigin-RevId: acae00e0a5

GitOrigin-RevId: 92307dd2ca

GitOrigin-RevId: 843dc3a790

GitOrigin-RevId: ff8eb003c5

GitOrigin-RevId: 7e6ea13be7

GitOrigin-RevId: 3c0690bcc1

as some reason: some version of trt/cudnn need
-Wl,--whole-archive attribute to fix cuda issue,
target A(CMAKE OBJECT,eg megbrain/megdnn): -Wl,--whole-archive depend on trt/cudnn
target B(SHARED LIBS,eg libmegengine.so) --> A with `PUBLIC` depends
target C(SHARED LIBS,eg _impertive.so) --> B with `PUBLIC` depends,
then ld will force link fatbin section into C, which will be undesired

what`s more, attribute PUBLIC/PRIVATE do not take effect
to OBJECT library(megbrain/megdnn)

what`s more, megengine/megengine_export have to PUBLIC for
mgb/imperative target, as SRC/include depends

so we pull cudalib depend from megbrain/megdnn to
megengine/megengine_export on linux os, to finall
target at windows os, for example lar on windows

GitOrigin-RevId: b278a69e1c

GitOrigin-RevId: 81e32da034

GitOrigin-RevId: 64d572bdfc

GitOrigin-RevId: aa4051541d

GitOrigin-RevId: 6f797570b6

GitOrigin-RevId: 4edc38eaf2

GitOrigin-RevId: d789c99e59

GitOrigin-RevId: 81c147aa6c

GitOrigin-RevId: 0fa3a534af

GitOrigin-RevId: 3b58b4acd9

GitOrigin-RevId: 2df8b6c164

GitOrigin-RevId: 337d95c7c2

GitOrigin-RevId: 9d83a174fa

GitOrigin-RevId: abfee3d4fa

GitOrigin-RevId: 372c37cdc5

GitOrigin-RevId: 8f56c43d1e

GitOrigin-RevId: 17a0cf6293

GitOrigin-RevId: ba251f452a

- refactor(mge): add support for optimizer.step().clear_grad() idiom

- refactor(mge): rename some methods of GradManager

- refactor(mge): remove tensor_nn and TensorDict

- refactor(mge): remove Buffer

- refactor(mge): remove requires_grad flag

- refactor(mge): add a default grad=None attribute to Tensor

- refactor(mge): deprecation for 1.0

GitOrigin-RevId: 3b723d9387

GitOrigin-RevId: 71ae7f1f4a

GitOrigin-RevId: 32429d8543

GitOrigin-RevId: f45a84696a

GitOrigin-RevId: 92b2c07bf9

GitOrigin-RevId: cb9ddcea3c

GitOrigin-RevId: 5e9a045410

1: fix cmake cross-ios failed caused by df118a87
   build static lib for APPLE define for XCODE third_party framework including require

2: megbrain_test/megdnn_test build when MGE_INFERENCE_ONLY=ON
   now u can build megbrain_test/megdnn_test by:
   EXTRA_CMAKE_ARGS="-DMGE_WITH_TEST=ON" ./scripts/cmake-build/xxx.sh
   example macos-cross-ios build megdnn_test for IOS by
   EXTRA_CMAKE_ARGS="-DMGE_WITH_TEST=ON" ./scripts/cmake-build/cross_build_ios_arm_inference.sh

3: reuse host flatc build when cross build mode
GitOrigin-RevId: 132f4bf893567bdb1d54de506449950513a5841f

GitOrigin-RevId: 7ffd9b9541

GitOrigin-RevId: 350ca7d15a

GitOrigin-RevId: 5c1992fb37

GitOrigin-RevId: 870d90900d

Also, upgrade to CMake 3.13.

The commit also contains significant refactors, as otherwise it is not
possible to properly export target `megengine` to
MegEngine-targets.cmake:

1. Optionally use system provided Flatbuffers.
2. Optionally use system provided MKL-DNN (Tested with Debian).
3. Refactor megbrain and megdnn targets into object libraries.
4. Set different path in BUILD_INTERFACE and INSTALL_INTERFACE of
   various target_include_directories.
5. Specify PUBLIC/PRIVATE on various target_link_libraries.

GitOrigin-RevId: df118a879e

GitOrigin-RevId: f341bea40b

GitOrigin-RevId: a2d9fa067a

GitOrigin-RevId: bfe5420b1d

GitOrigin-RevId: aca796f17d

GitOrigin-RevId: f870ad964c

GitOrigin-RevId: 26a8dc50b8