Feature/make slit illum interactive

[olesjas]

list of changes:

• Added several parameters to the makeSlitIllum primitive, mostly following IRAF task example. E.g. the possibility to supply bins as list of regions, the option to specify sample regions along the spatial axis, interactive reduction.
• Replaced the final step of creating slit function from fitting a 2D model to 1D row interpolation.
• Two interactive interfaces were created for makeSlitIllum, for dispersion bin inspection and editing, and for dispersion bin fitting. For the first interface I had to make a specialized visualizer to be able to use regions interface without having to deal with all the fitting-related parts (bineditor.py).
• By default, the extrapolated parts of slit function are being set to a constant value equal to that of the last interpolation point (as it is done in IRAF), but it can be set to use extrapolation.

[olyoberdorf]

Looks great - thanks!

[chris-simpson]

I've pushed a number of commits to feature/makeSlitIllum_reviewing -- I created a new branch because I changed the way the weights were calculated when collapsing the bins and thought this might cause regression tests to fail. It's also recently had master merged into it. Some are small changes to the calculations, some are simply stylistic to shorten the code (which I think is always good, as long as it doesn't significantly affect the readability).

I would also favour removing the logging statements as I don't think they really add anything. The level of detail in the logs varies a lot from primitive to primitive but generally I would suggest a logging statement for the user whenever something slow is about to happen, or when a non-standard path is taken. Logging should also be used (perhaps at the debug level) when numbers are read from files (to make sure the correct number is being read) or when an if statement causes a decision to be made, to aid with debugging.