Disable compatibility mode if intensity is a multiple of worksize. In that case enabled compaibility mode will only slow down the miner.

If two threads are using the same GPU device the start time of each hash round is optimized based on the average time needed to calculate a bunch of hashes.

This way to optimize the hash rate was first introduced by @SChernykh. This implementation based on the implementation in xmrig but differen in the details.

- introduce a new config option `interleave`
- implement thread interleaving

Add new striding index where the memory is chunked by the size of the work group (worksize).

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

- change a few 64bit variables into 32bit.
- provide defines type quallified

- remove useless `clFinish`
- avoid download num threads for skein&co and start always as much threads as in all other kernel (terminate useless threads)

- optimize kernel cn0 and cn2
- optimize vast int math
- use more 32bit variables

Co-authored-by: psychocrypt <psychocryptHPC@gmail.com>

optimize the devision in cryptonight_heavy and cryptonight_haven

import of https://github.com/xmrig/xmrig-amd/pull/185/commits/5d9b9334654df25cea7707f667990fd1577ed290

The AMD compiler for OpenCL shipped with the driver 14XX is broken
and can not compile xmr-stak since the monero v8 changes are introduced.

- workaround a simple compare.
- add new device define `OPENCL_DRIVER_MAJOR`

Strided index 1 is not allowed for cryptonight_v8 and monero.
In the case the dev pool is set to monero and the user tuned there settings for
an other currency the miner will crash if strided index or memChunk is not
fitting the requirement to mine monero.
This PR detects wrong configurations and will set strided index and memChunk to a valid
value but only for cryptonight_v8. The user pool settings will only be changed if monero or
cryptonight_v8 is selected.

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

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

- 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>

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.

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

- add cryptonight_heavy derivate cryptonight_bittube2
- add coin bittube
- remove coin ipbc because this coin is now called bittube

- rename cryptonight_fast to cryptonight_masari
- set dev pool to cryptonight_monero

solve #1494

- add algorithm `cryptonight_v7_stellite` (internal named: `cryptonight_stellite`)

- add algorithm `cryptonight_lite_v7_xor`
- update documentation

- fix that to much OpenCL calls whre used if cache is disabled

remove fork version for sumokoin and monero7

- allow the dev pool to fork on a different block version than the user descriped coin

All algorithm are centered around the user coin description.
It is allowed to have two two different coin algorithms in the user coin description.
It is only allowed to use algorithms for the dev pool coin description those are used in the user coin description.
There are two ways to define a non forking coin.
  - set both user coin algorithm descriptions to the same algorithm and set version to zero
  - set the first algorithm in the user coin description to something you like to use in the dev pool
    and set the second algorithm to the correct representation of the coin. Set the version to 255.
    This will allow that the dev pool can mine on a different coin algorithm than the not forking user coin.
    Do not use an algorithm with different scratchpad size for the dev pool.

- add CLI flag to explicitly use non AMD OpenCL and devices
- adjust OpenCL output (use OpenCL instead of AMD if --altOpenCL is sued)
- optimize NVIDIA OpenCL auto suggestion

Use the maximum scratchpad size from before and after the fork.

- remove version numbers within the kernel
- create seperate program context for each mining algorithm
- remove kernel `cn1_monero` is now integrated in `cn1`
- remname `cnX` kernel in `cnX + algorithmNumber`

- fix that version argument was not passed to extended kernel parameters

- add `fork_height` to currency
- refactor algorithm selection

allow to disable the OpenCl cache
  - usefull for read only systems
  - usefull for unknown errors during cache reading

fix warning

```
/Users/user/xmr-stak/xmrstak/backend/amd/amd_gpu/gpu.cpp:481:11: warning: using the result of an assignment as a condition without parentheses [-Wparentheses]
                if( ret = clGetProgramInfo(ctx->Program, CL_PROGRAM_BINARIES, num_devices * sizeof(char*), all_programs.data(),NULL) != CL_SUCCESS)
```

- add new pow for AEON
- fix missing cryptonight-heavy selection for multi hashes

change warning when precompiled OpenCL code is not found

Co-authored-by: psychocrypt <psychocryptHPC@gmail.com>
Co-authored-by: fireice-uk <fireice-uk@users.noreply.github.com>
Co-authored-by: Lee Clagett <code@leeclagett.com>
Co-authored-by: curie-kief <curie-kief@users.noreply.github.com>

A redefinition of a variable in a local scope avoid that the intensity is rounded to a multiple of the work size.

Reduce OpenCL start time by using a self made compiler cache.

- store compiled OpenCL binary
- load OpenCl binary if available

fix double definition of define `MEM_CHUNK`