One reason you may not find anything as simple as an --auto option for VARs is that data mining of this type (setting zero restrictions) is counter to the VAR modelling philosophy. Also it is rarely necessary because setting lag coefficients to exactly zero when the OLS estimates are statistically similar to zero will typically have few statistically discernible effects in the performance of the entire system (that's why they are statistically insignificant).
GRETL does have a VAR lag selection tool that does not violate the general VAR model building concept, but this is probably different from what you desire. Lasso in R might be the most reasonable alternative, especially if you are using the VAR to forecast.
Or consider Bayesian VAR modeling that lets you shrink the lagged coefficients toward around priors that are centered around zero. Check out Nobel prize winner Chris Sims R code to implement (http://www.princeton.edu/~sims/#VARtools)