NVIDIA: fix wrong number of threads

In the cuda backend for monero we start always twice as much threads as needed.
Those threads are than removed after the AES matrix is copied to the shared memory.
Never the less it is the result of an copy past bug.

- start correct number of threads for `monero`

xmrstak/backend/nvidia/nvcc_code/cuda_core.cu

@@ -734,7 +734,7 @@ void cryptonight_core_gpu_hash(nvid_ctx* ctx, uint32_t nonce)
 {
 	dim3 grid( ctx->device_blocks );
 	dim3 block( ctx->device_threads );
-	dim3 block2( ctx->device_threads << 2 );
+	dim3 block2( ctx->device_threads << 1 );
 	dim3 block4( ctx->device_threads << 2 );
 	dim3 block8( ctx->device_threads << 3 );