recursive feature elimination with caret

model-free variable importance

Fix caret tuning, and plot var importance

fix VL bug, and use lag to shift local variables so for every trial we only consider previous local params.

fix param

remove current trial from local stat calculations.

reactor, and add plots

a new glmnet (lasso and ridge) and pls analysis with dummy vars for stimulus, stimulus_type, and N.

init a new dummy var analysis for ccn2019