Print residue with insertion code

[smoe]

This was a bit of a technical exercise while continuing the thoughts expressed in issue #544 on the visibility of the insertion code. The getFullName method is extended respectively and the python_toString routine falls back to it. The patch required to also update and extend the FullnameType enum for Python.

[dstoeckel]

I think this looks pretty good. One suggestion I would like to make though is to use the FullNameType enum as bitflags that are used to build the combined flags:

enum FullNameType {
    NO_VARIANT_EXTENSIONS = 0,
    ADD_VARIANT_EXTENSIONS = 1,
    ADD_RESIDUE_ID = 2,
    // ADD_VARIANT_EXTENSIONS | ADD_RESIDUE_ID
    ADD_VARIANT_EXTENSIONS_AND_ID = 3,
    ADD_INSERTION_CODE = 4,
    // ADD_INSERTION_CODE | ADD_RESIDUE_ID
    ADD_RESIDUE_ID_AND_INSERTION_CODE = 5,
    // ADD_INSERTION_CODE | ADD_RESIDUE_ID | ADD_VARIANT_EXTENSIONS
    ADD_VARIANT_EXTENSIONS_AND_ID_AND_INSERTION_CODE = 6
}

The combined flags can stay for convenience (and API compatibility), the checks for building the full name would be a lot easier to read though...

@anhi What would you say about that proposal?

[smoe]

An ABI incompatibility sounds like a close-to-perfect motivation to perform a library jump from 1.4 to 1.5 :) I was just saying that to satisfy my ambitions for Debian. I have not fully grasped what exactly you were commenting, but if manually assigning values, then the value of ADD_INSERTION_CODE | ADD_RESIDUE_ID | ADD_VARIANT_EXTENSIONS I propose to be 7, not 6 - should also help simplifying the implementation quite a bit.