Consider switching to SI units

[myurkin]

Currently ADDA uses CGS units for all formulae (including the manual) and input and output quantities. This corresponds to earlier reviews of the DDA (e.g. Yurkin & Hoekstra 2007). In most cases the choice of units is not important, since most measured quantities are either dimensionless or have units of area. However, the issue becomes less trivial for normalizing of input power of non-plane beams and for radiation forces (#132 , #14 ). In discussion of those issues the users seemed to be more comfortable with SI units, especially when considering a real experimental system.

Moreover, more recent papers, relevant to the DDA use SI units. Hence, it may be less convenient to implement new ideas into ADDA, if the latter would stay in CGS. On the other hand, switching to SI may have some back-compatibility problems. Apart from the output quantities described above, it may change, e.g., the dipole polarizations (can be saved for processing by other utils, such as near_field package).

[myurkin]

After looking at how the DDA formulae transforms between those two system of units, it seems reasonable to express dipole polarization P and polarizability α in units of ε0 (both inside the code and in output files). There will still be difference by a factor of 4pi.