Overhaul functions in statistics/base directory. Widen input validation to accept logicals. Return correct class of output, e.g., 'single' depending on class of input. Correct or add tests for above. * center.m, cov.m, kendall.m, mean.m, meansq.m, median.m, mode.m, prctile.m, quantile.m, ranks.m, run_count.m, runlength.m, spearman.m, statistics.m, std.m, var.m, logistic_inv.m: Overhaul as described above * corrcoef.m: Overhaul + remove input validation already done by cov(). * cor.m, logit.m, ppplot.m, table.m: Only align test blocks. * gls.m, ols.m: Only correct class of output, no logical inputs for regression. * histc.m: Only change spacing of code to be uniform. * iqr.m: Overhaul + 2X speedup by calling empirical_inv just once. * kurtosis.m: Overhaul + replace repmat instances with center(). * mahalanobis.m: Overhaul + use bsxfun for centering data. * moment.m: Overhaul + replace repmat instances with center(). * probit.m, range.m: Redo input validation and add tests. * skewness.m: Overhaul + replace repmat instances with center(). * zscore.m: Overhaul + replace repmat instances with center() + use bsxfun.