Allow more variation of the formula used for amplitude in LauFlatteRes

[Thomas Latham]

In the literature the exact formula used for the Flatte lineshape varies quite a bit. In some cases the m_0 factor appears in the imaginary part, in some cases it does not. So we should use a flag to determine whether or not that appears.

The values used for the a0(980) from PRD 57 (1998) 3860 are incorrect. Firstly, the value we take for g1 is not their value but another, it should be 324 +/- 15 MeV. Secondly, their formula uses g^{2 rather than g, so we need to then square that value. The ratio is of the g}2 values, however, so no need to change that.

Should perhaps also set the mass values in the constructor as well as the g values.

Probably should also add some printout to confirm the settings being used.

[Thomas Latham]

(In [532]) Correct LauFlatteRes to use the exact formulae from the papers that report the default parameter values for each supported resonance, fix #77

• Deals correctly now with the cases where the m0 is absorbed into the couplings
• Now uses the correct value of g1 for the a_0(980) (needed to be squared)
• Also sets the mass values to those reported in those papers
• Added printout to specify what is being done in each case