Store both coefficients and amplitudes as attributes

[Schoyen]

So far we have distinguished between methods with static orbitals (TDCC) and time-dependent orbitals (OATDCC) in the time-evolving case. However, as seen in the OACCD ground state calculation we need both a set of amplitudes and a set of coefficients. This will also be the case for RCC2 with the t_1-transformed system (#58). We should treat the coefficients and the amplitudes on an equal footing as in the time-dependent case.