scikit-learn: R-like summary of LogisticRegression

It would be very nice to have R-like summary() function. This is a popular question. I understand it doesn’t make sense to output R^2 by default along with everything else. But I think some summary is nice.

My proposal: When instantiating a model, allow the user to specify a header column. This will be used as the header column so that .coef_ can be outputted as a Pandas series, or another function can be added which would output this. This would make debugging models significantly easier.

Are there major objections or should I just implement it and submit a pull request?