updating scatter

Following on from #44, there's a number of things that should be improved.

Easy enough:

• more documentation
• downsampling on activity and attenuation image should be made default in ScatterSimulation
• enabling of FBP recons (for scanners without gaps)
• rename some classes to include the PET name (as wouldn't work for SPECT etc)
• add a simple test for ScatterEstimation (fixed in #737)
• don't hardwire name and location of ./extras folder for extra ("debugging") output
• add an option for the image output during iterations (currently written in ./extras if debug is on) [postponed until somebody asks for it]
• ScatterSimulation should be able to upsample directly

Harder:

• use of ChainedBinNormalisation is counter-intuitive to me (fixed in #744) (I think we should be able to not give the ACFs)
• use of the tail-fitting parameters for ACFs is weird when constructing it from an image mask
• alternative tailfits, e.g. without having to subtract randoms in the code
• avoid rerunning the sensitivity calculation for OSEM etc, see #476
• allow ScatterSimulation::set_up to be called twice. Currently it would downsample twice. We try to prevent this from happening but it can still fail.

Hard:

• upsampling of non-oblique segments, required for long scanners
• TOF

Start by splitting the remaining requests into separate issues and reading the ScatterSimulation setup path, the ScatterEstimation and ChainedBinNormalisation references, and related issue #476. The already-closed items and postponed image-output request should not be treated as current scope; each remaining change needs its own acceptance criteria before work can be considered done.