Support different sampling modes in segments.py

[kahst]

It would be nice to have two sampling modes in segments.py. The way it's implemented now is to randomly shuffle all segments for a species and then limit the number of returned segments to max_segmens (see here) Additionally, we should provide a way to pull a predefined number of segments per scoring range (e.g., 10 segments for scores in [0.5, 0.6[ and [0.6, 0.7[ and [0.7, 0.8[ and so on).
Should be a command line option called uniform and represenatative or similar.