No update to posterior of psi for A > 1

In the example repo's [example notebook](https://github.com/facebookresearch/clara/blob/master/examples.ipynb), it is clear that when A > 1, the posterior of psi is only updated for labeler of index = 0, the rest of which does not get updated.

This seems to be because when `self._init(ratings, labelers, true_ratings, scores)` is called in` ClaraGibbs.fit()`, initializing assignment only works for the label of index = 0, the `data` field was not updated for other labeler indices.

I managed to replicate the same problems using my own dataset.

Start with examples.ipynb and trace ClaraGibbs.fit(), especially the call to self._init(ratings, labelers, true_ratings, scores). Compare how the data field is initialized for labeler index 0 versus other labeler indices when A > 1, then verify that the posterior of psi updates for every labeler using the example or the reporter’s replicated dataset.

Written by the indexing model from the issue text.