Backport transposes optimization to v0.3.0 (pytorch#3994)

* Optimizer: optimize transposes in variety of circumstances (pytorch#3509)

* Optimizer: Optimize transposes in variety of circumstances

- No-op transposes
- Consecutive transposes (fuse them)
- Transposes into Gemm (fuse them into transA/transB parameter)

* touch up out of date comment

* Backporting optimizer changes

fix static linkage and make THD statically linked

version bump

check for nvidia driver's sufficiency before checking for number of C…

…UDA devices (pytorch#1156)

import numpy before setting dlopen flags (pytorch#928)

Fix Engine::compute_dependencies

std::move fixes

add cc 3.0 to nccl (pytorch#594)

fix docs for torch.nn.functional.conv1d (pytorch#536)

tensor docs