unify cpu cryptonight implementations

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.

my speed changes

now it is fast

human readable

xmrstak/backend/amd/minethd.cpp

@@ -252,7 +252,7 @@ void minethd::work_main()
 
 				*(uint32_t*)(bWorkBlob + 39) = results[i];
 
-				hash_fun(bWorkBlob, oWork.iWorkSize, bResult, cpu_ctx);
+				hash_fun(bWorkBlob, oWork.iWorkSize, bResult, &cpu_ctx);
 				if ( (*((uint64_t*)(bResult + 24))) < oWork.iTarget)
 					executor::inst()->push_event(ex_event(job_result(oWork.sJobID, results[i], bResult, iThreadNo, miner_algo), oWork.iPoolId));
 				else

xmrstak/backend/amd/minethd.hpp

@@ -24,7 +24,7 @@ class minethd  : public iBackend
 	static bool init_gpus();
 
 private:
-	typedef void (*cn_hash_fun)(const void*, size_t, void*, cryptonight_ctx*);
+	typedef void (*cn_hash_fun)(const void*, size_t, void*, cryptonight_ctx**);
 
 	minethd(miner_work& pWork, size_t iNo, GpuContext* ctx, const jconf::thd_cfg cfg);