Reinterpret a complex-valued floating-point array as a real-valued floating-point array having the same precision.
