Parallel averaging

[lindonroberts]

trafficstars

Currently, if nsamples>1, objfun is called multiple times in a loop. Allow for extended objfun evaluation

objfun(x, nsamples)

where the objfun calculation returns all nsamples results in one go (instead of a for loop inside this code). This allows for things like parallel evaluations, if objfun supports this (e.g. MHD simulations of @florianwechsung).