Clarify reference metallicity for SSE models and winds

[ilyamandel]

ZSOL in constants.h refers to the Hurley models, and should really be called ZREFERENCEHURLEY or some such, not ZSOL, to avoid confusion. For winds, we should go back to the individual papers — e.g., the WR scaling according to Vink & De Kotter (2005) is a relative scaling with metallicity, so the exact value of ZSOL doesn’t matter. Of course, it does matter for the prefactor — which, in Vink+ 2001, is taken as Zsol=0.019. So, if we want to fairly use these wind models, we should scale them to whatever Zsol was assumed there. [For WR winds, we already have a free multiplicative factor, so I am not too worried about those.]. Anyway, sounds like we should add an issue to go through our wind models and check that we are using the appropriate reference metallicity for each one — and do away with the ambiguous ZSOL.

[jeffriley]

ZSOL defined in constants.h is also the default metallicity used for SSE and BSE - used if the user does not specify metallicity via a program option or grid file.

[ilyamandel]

Reassigning given the relevance for winds.

[ilyamandel]

@SimonStevenson and @jmerritt1 -- any thoughts on this one?