1-d MCMC outputs

[florianhartig]

Based on #221 - check that all functions also work for 1-d MCMC outputs

[RobertBosek]

after fixing matrix subsetting in marginalLikelihood.R and plotMarginals.R functions should work for one parameter MCMC-samplers

tested with:

ll <- function(x) sum(dnorm(x, log = TRUE))
setup <- createBayesianSetup(ll, lower = -100, upper = 100)
out <- runMCMC(bayesianSetup = setup, sampler = x, settings = list(iterations=1000))

used "Metropolis", "AM", "DR", "DRAM", "DE", "DEzs", "DREAM", "DREAMzs", "Twalk" samplers as x

tested functions print, getSample, summary, plot, DIC, MAP, getVolume, marginalLikelihood, marginalPlot, tracePlot work for all samplers

WAIC, gelmanDiagnostics also work for samplers with multiple internal chains ("DE", "DEzs", "DREAM", "DREAMzs", "Twalk")

correlationPlot correctly exits with warning message:

Error in correlationPlot(out) : 
  BayesianTools::correlationPlot - using this function only makes sense if you have more than 1 parameter