Compilation errors with single precision enabled

Created originally on Bitbucket by ofmla (oscar mojica)

I am getting errors like this one

‌

fft3d_with_fftw3d.cpp:49:10: error: cannot convert ‘float ()[2]’ to ‘mycomplex {aka std::complex<float>*}’ in assignmentarrayK = fftwf_alloc_complex(nK0 * nK1 * nK2);

when compile the test_bench_cpp code using SINGLE_PREC:=yes in the Makefile. You also should use reintepret_cast for single precision case

‌