Pull request for steplike isobars.

[fkrinner]

steplike isobars implemented, which can be created from a single keyFile

[legordian]

lorentzFactor change: b1c3876ab9ed4e0a570f20e5a195a166ed8f7abd

[suhlatwork]

A couple of things are not working with the latest state of this request:

• different waveDescriptions expanded from the same keyfile get the same name which is a bad thing for amplitude file names, ...
• the changes to the file manager are not in agreement with the changes done on the C++ side

I already did some changes to the code I agree with, mostly those have been cosmetic (white-spaces and other coding-style issues) except for:

• your step-like mass dependence would always return amplitude 0 for a mass directly on the bin border, now in such cases the amplitude would be 1 in the higher bin
• the while (true) loop during the expansion has been replaced with recursion (which I think makes the code a bit more readable)
• get and mapMassDependenceType have been merged into one again

Please carefully check and continue working based on e544f47da9418c195e9e1c17f9017d32d47ff847.

I am still not happy with:

• the flatRangeMassDependence should be removed now that this better way becomes available
• the name steplikeMassDependence of the massDependence class (a step only goes either up or down), I would prefer rectangularMassDependence or re-using flatRange...
• the name of the mass-dependence given in the keyfile (now freed-isobar/freedIsobar, but strictly speaking a single bin is not a freed-isobar), as for the other mass-dependencies we should use the same name here as for the class

Let us continue working on this once you are back.