check gpu architecture 

AMD: increase memory pool

fix wrong memory detection

* CONF_NO_TLS fix

* gpu error fix + always return to sched after pin

- check if the gpu architecture is supported by the compiled miner binary
- remove not supported gpus from the auto suggestion
- disallow the selection of a not supported gpu by hand tuning the config

Free and total memory is only evaluated on the first device.
To detect the gpu memory the gpu must be selected.

- create context on the gpu before the memory is checked
- add smx to the auto detection
- change the result code of `cuda_get_deviceinfo()`

Documentation: Add environment variables to increase OpenCL memory pool

Versioning

- update usage.md
- hpdate help output

Change attribution

Added long version as a separate option

improve cli hash report

native compile by default

fix wrong box ending

Add version to web and json

- fix wrong text box

- add CMake flag `XMRSTAK_COMPILE` to select `native` or `static` compile
- add new flags to the documentation

- generate define `BACKEND_TYPE` type in CMake
- fix comile because `GIT_COMMIT_HASH` is no string

Split hashrate report into backends

- add type of the backend to each backend-plugin
- add `gteName` to `iBackend` to get the name of the backend

Versioning final version

Define GIT_COMMIT_HASH via cmake or git archive

* Multi-pool first draft

* Fix wspace from new IDE

* Better TLS error message

* Fix TLS bug

* Don't put dev pool on stats + pool change-back

* bug fixes

* Error message work

* fix win build

* add per-pool nicehash setting

* Fix bugs

* rm debug msg

* Multipool guided setup

* Support TLS and Nicehash in config

* prelim jconf changes

* final multipool changes

* increase default retry_time to 30, fix mac erro

* rm debug dev pool settings

* Fix another source of connect runaway

update compile documentation

- add CUDA and AMD APP SDK dependency to linux compile guide
- add AMD APP SDK dependency to windows compile guide

fix windows CUDA 9 compile

fix #91

- add CMake workaround for wrong host compiler for cuda
- update winows compile documentation

optimize NVIDIA autosuggestion

- avoid creation of a config with zero threads or blocks
- WINDOWS: reduce the used memory for the auto suggestion by the amount of already used memory

search for linux backend plugins

- search for plugin in
  - working directory
  - binary directory
  - let linux search the plugins (libraries)

OSX file file extension (dynamic libraries)

fix OSX compile

fix `/usr/bin/ld: bin/libxmr-stak-backend.a(jconf.cpp.o): relocation R_X86_64_32 against `.rodata.str1.1' can not be used when making a shared object; recompile with -fPIC`

use `.dylib` as dynamic library file extension