This is an alternative implementation of a subset of the TVM runtime API (and
graph runtime) that focuses entirely on reducing code size, at the expense of
functionality (no tvm.extern(..) calls via PackedFunc, CPU only, etc). It might
be worth incrementally expanding the surface area if there's interest.

The motivation for this work was seeing what the minimal useful subset of the
TVM runtime is. This is relevant for e.g. super code-size constrained
applications in e.g. embedded/mobile. The current runtime is more like O(100KiB)
or so, so this might be compelling for some users.

The smaller surface area for auditing might make this relevant for
https://github.com/dmlc/tvm/issues/3159, or the usecases I was thinking about in
https://github.com/dmlc/tvm/issues/2523#issuecomment-459165815 re: the Rust
runtime.

The symbols in the tvm::minimalruntime space (i.e. excluding std:: and
picojson::) are about 5KiB, so I think there's a bunch of room here (i.e. we
could replace picojson:: with [`jsmn`](https://zserge.com/jsmn.html) or
something, and we could replace more of the `std::unordered_map` usage, etc with
custom primitives as well (similar to the `DynArray`).

* Module refactor

* Add load module

* Add support for idempotent import

* Tweak load paths

* Move path around

* Expose C++ import functions in Python

* Fix import

* Add doc string

* Fix

* Fix lint

* Fix lint

* Fix test failure

* Add type solver

* Fix lint

* support LLVM trunk

* guard with USE_LLVM in if condition for c++14

* GREATER_EQUAL -> GREATER

* [Arm] parallel batch axis

* [Relay][Frontend][Keras] Fix ReLU in Keras Converter missed the case

* [Relay][Frontend][Keras] Add test case for ReLU in Keras Converter missed the case

* [Relay][Frontend][Keras] Add test case for ReLU in Keras Converter missed the case

* save

* save

* numpy compatible type inference

* update

* try to fix

* fix

* try to fix

* fix lint

* Update nn.h

* cast to int32

* try to fix

* fix again

* retrigger ci

* add more ops

* stop vectorization for erf

* x

* cleanup

* fix

* add whitelist for vectorizable intrin

* add tf converter

* fix dense

* fix

* add missing intrin

* fix mxnet frontend

* fix nvptx

save

fix

fix grad

* fix cmake for mac os

* rename

* support LLVM trunk

* guard with USE_LLVM in if condition for c++14

* GREATER_EQUAL -> GREATER

fix lld

* [VTA] Support TLPP in function simulator.
Issue:
currently vta function simulator just doing serialized instruction
execution, the dependency logic of runtime ISA which use for task
level pipe line parallelism can not get verified by function simulator.

Solution:
make the simulator driver to be multiple thread and support TLPP.

Benefit:
TLPP support VTA function simulator would make VTA logic testing/debug
/change more easy.

replace boost lockfree queue

add configure control for simulator tlpp enable or disable.

change code tyle into google style.

Wrap queue read/write and sync logic to make function call more simple.

Add some comments.

Remove MT logic, change into Single thread mode.

address review comments.

code style change to match google code style and add comments.

add cmake macro to enable/disable simulator tlpp logic.

submodule update.

correct file name mentioned in comments.

* remove USE_VTA_FSIM_TLPP.

* update lint

* lint fixed

* lint updated

* lint fixed

* lint fixed

* lint fixed

* updates

* add intel graphics as a package

* remove print info

* depthwise conv2d schedule added for intel graphics

* asdf

* fix lint

* fix lint

* fix ci

* add channels

* save

* init

* move type_relations

* Getting closer to having ADT defs

* ADT defs working probly

* Match parsing basipally done

* came to earth in a silver chrome UFO

* match finished?

* All tests but newest are passing

* ADT constructors work

now cleanup?

* Cleanup round 1

* Cleanup round 2

* Cleanup round 3

* Cleanup round 4

* Cleanup round 6

* Cleanup round 7

* Lil grammar fix

* Remove ANTLR Java files

* Lint roller

* Lint roller

* Address feedback

* Test completeness in match test

* Remove unused imports

* Lint roller

* Switch to Rust-style ADT syntax

* Lil fix

* Add dummy `extern type` handler

* Add type arg to test

* Update prelude semantic version

* Repair test

* Fix graph var handling in match

* Revert 's/graph_equal/is_unifiable' change

Installed through pypi

* Search PATH for DLLs

* Fix lint issue

* implement tuple pattern

* add tuple pattern

* lint;

* lint

* lint

* fix error

* fix

* add test

* save

* add test

* refactor

* fix indent

* save

* refactor

* Update from_source.rst

* Update deploy_ssd_gluoncv.py

* [Relay] enable NHWC of `relay.testing.mobilenet`

In this way, we can play around NHWC inside TVM regardless of
the frontends.

* [Test] test for NHWC of relay.testing.mobilenet

* initial conv2d_transpose

* correct select operator

* cleanup

* fix

* fix correcness check

* conv2d transpose declaration fix

* autotvm conv2d_transpose tuning script

* ir pass fix

* fix tuning script

* deriving params from env, adding bias

* removing bias comp from deconvolution

* lint

* fix

* lint

* lint

* turning off cpu

* lint, ops

* lint

* import fix

* removing hard coded values

* lint

* adding support for graphpack over multiply op

* increasing resnet model coverage

* fix indentation

* lint

* moving recursion limit fix into graphpack pass

* moving recursionlimit to relay init

* pooling on NCHWnc format

* adding more models

* deploy_resnet_on_vta.py

* trailing line

* generalizing to vision models

* merge conflicts

* fix, apply quantization to VTA only

* improving comments

* trimming models that have runtime issues for the moment

* lint

* lint

* lint

* init

* fix

* rework;

* `de10-nano` -> `de10nano`;

* fix compilation error;

* bug fix;

* Update install.md

* Update install.md

* Update install.md

* update with current runtime;

* add debug messages;

* bug fix in cma kernel module;