Apply premature stopping features to iterative algorithms.

[geektoni]

This is a good entrance task for GSoC's students who want to tackle the Inside that black box project. This issue is good also to understand how Shogun's internals works and how we generate the cookbook.

First of all, take an algorithm which is trained iteratively (for example, CPerceptron) and do the following:

• Propose/Discuss which are the suitable actions that can be performed when pausing/stopping the model (what are the information we want to show to the user, etc.);
• Implement the corresponding methods (on_pause(), on_complete(), on_next());
• Add a meta example which shows these new features such that a prospective user can see how premature stopping works;

These changes ideally should be done in separate PRs. See also the links below for some detailed information about how the premature stopping architecture is structured:

• Premature Stopping Guide

[prashanthduvvada]

Will take up the issue