Added variant ions and more complex satellite ion definitions to the theoretical model

[douweschulte]

Variant ions (z/z·/z') are quite common, especially in ETD and UVPD. While researching these I also found that w ions can be formed away from the main z ion backbone cleavage location. So the current model allows those fragments as well.

todo:

• Apply variant ions

[douweschulte]

@di-hardt this PR has hit the stage where so much has changed that I am not sure I dare ask you to review. I am nearing completion, with the last pieces of logic on glycan fragments on peptide fragments still to be written. To give a quick run doen of stuff in here:

• Glycan fragments on peptide fragments (so what happens to a z peptide with a glycan versus a y ion)
• Neutral losses of glycans and immonium ions
• Specific neutral losses based on amino acids and amino acid side chain losses on main series and precursor ions
• Satellite ion redefinition to allow secondary fragmentation at a distance
• 'Variant' ions ( vs z+1 vs z-1 etc)
• General bug fixes
• Exporting fragments to mzPAF (compared to the rest quite a simple addition)
• Redefinition of the theoretical models to use the above things when needed.

Let me know if you are still interested in reviewing. Sorry for making a mess of this PR but all the glycan stuff is really increasing the complexity.

[di-hardt]

Yeah sorry, I looked over your commits when they came in. Due to the complexity I wanted to do a full review this week. Anyway looked fine to me in the end.