Fixing 1D array deprecation warnings for model_params property

[desilinguist]

Currently, in model_params we are using the following to get the coefficients for the selected features for a linear model:

coef = self.feat_selector.inverse_transform(coef)[0]

This sometimes raises DeprecationWarnings of the following kind:

/Users/nmadnani/anaconda/envs/rsmtool/lib/python3.4/site-packages/sklearn/utils/validation.py:386: DeprecationWarning: Passing 1d arrays as data is deprecated in 0.17 and willraise ValueError in 0.19. Reshape your data either using X.reshape(-1, 1) if your data has a single feature or X.reshape(1, -1) if it contains a single sample.

This happens because the inverse_transform() method of the feature selector requires a 2D array of shape [n_samples, n_features]. So, to get around these warnings we should simply change the SKLL code to be:

coef = coef.reshape(1, -1)
coef = self.feat_selector.inverse_transform(coef)[0]

[desilinguist]

Addressed by #329.