Read memory in bigger chunks per thread to increase the used memory bandwith.
Use for Kepla and Fermi GPUs the old autosuggestion instead of the new settings for cryptonight_v8.

I disabled a few algorithms for fatser compile and missed to re-enable them.

- restructe asm preparation function
- add double hash asm code

`uint` is unknown in windows, therefore switch to the better type `uint32_t`

In the auto adjust without hwlock the asm entry was missing

- introduce a new schema where two threads work together on one hash
- update autoadjustment
- remove an mistake where shared memory was shrinked for gpus < sm_70

apply optimizations

Co-authored-by: SChernykh <sergey.v.chernykh@gmail.com>

add cpu implementation for the final monero POW

- remove unused host function (relict from old refactoring)
- remove unused OpenCL full div function

Avoid branche differegence

- fix that shared memory for fast div is always used even if an algorithm is not using it
- optimize fast div algo
- store `division_result` (64_bit) per thread instead of shuffle around and store it as 32bit

- use optimzed div and sqrt
- reduce memory footprint

Add fast version for div and sqrt for the cuda backend

- fix code style issues
- fix spelling issue
- fix asm to support newer clang versions

- reintroduce monero7 until the POW is final
- update docs (add cryptonigh_v8)

- add special asm version for win64 and linux
- add cmake path for MSVC and other systems

- fix assembler code to pass the clang compiler
- CMake: set asm file language
- fix icc with gcc-7 compile issue with `_addcarry_u64`

Remove the asm option `auto` by `off`

If single hash is used the type of the variable to hold the intermediat sqrt value is
changed from `__m128i` to `uint64_t` as suggested by @SChernykh

- add new option to `cpu.txt` named `asm` to select the asm code version
- extent function selection method to choose assembler code for `cryptonight_v8`
- update auto adjustment to add default value for option `asm`

Add @SChernykh assembler version for ryzen and intel processors.

Co-authored-by: SChernykh <sergey.v.chernykh@gmail.com>

add option `unroll` for OpenCL to allow better tuning the main POW kernel.

Create a special pass for NVIDIA GPUs to load memory chunks first into the shared memory.

Co-authored-by: SChernykh <sergey.v.chernykh@gmail.com>

- use shared memory to exchange

implement `cryptonight_v8`

- implement cryptonight_v8
- update auto adjust to fit the special requirements of `cryptonight_v8`
- add fast math integer implementation for `sqrt`, `reciprocal`  and `division`

Co-authored-by: SChernykh <sergey.v.chernykh@gmail.com>

Add support for single hash cryptonight_v8.

Co-authored-by: SChernykh <sergey.v.chernykh@gmail.com>

- rmeove currency `monero7`
- introduce `cryptonight_v8` and `monero8`

During the initialization of the compile parameter for OpenCL it could be that the
fixed size buffer is to small. To avoid this we are now using `std::string`.
There is no problem by using `std::string` because this part of code is not perfromance critical.

If the first bit of the nonce is `1` (this is very often if we use a nicehash pool)
than it could be that some OpenCL implementations handle the 64bit representation of the 32bit
nonce on the device side as signed integer.
During a right bitshift we pull wrong ones from the wrong higher part of the 64bit
nonce representation into the 32bit part of the nonce.
The result will be that the computed share is invalid.

- explicit cast the nonce on the device to `uint` to avoid any side effects

Avoid that a OpenCL binary from the cache is used if the driver or xmr-stak version has changed.

There is a copy past mistake tha tthe type of the variable `memChunk` is not tested.

xmr-stak has several implementations for multi hash per thread.
The results into 3 intepedent implementations.
Each time the algorithm must be changed the possibility to introduce errors is very large.

- unify the different cryptonight CPU implementations
- simplify the function selection array to find the specilized cryptonight implementation
- add a intermediat pointer to access the large state (similar to the old multi hash implementation)

As side effect this change increases the speed of the single and multi hash algorithm.