GMM compute summary and update distributions in one pass

In current impl, GMM need to trigger two jobs at one iteration:

1, one to compute summary;

2, if {{shouldDistributeGaussians}} ((k - 1.0) / k) * numFeatures > 25.0,

trigger another to update distributions;


 

shouldDistributeGaussians is almost true in practice, since numFeatures is likely to be greater than 25.

 

We can use only one job to impl above computation,