Update dependencies, available readers, wildcard selections

[lilyminium]

Fixes #66
Fixes #65
Fixes #55

Build: https://mdauserguide.readthedocs.io/en/update-stuff/

Changes:

• Update wildcard selection section to reference fnmatch
• Update dependencies needed to build MDAnalysis, moved doc dependencies to the same place
• Update topology / reader tables
  • NAMDBIN (new)
  • FHIAIMS (new)
  • CHEMFILES (new)
  • PARMED (new)
  • PDB has elements now
• Auto-generated tables now say which script generated them and include the table heading
• auto-generation on build still remains turned off for now

[lilyminium]

@Luthaf @Marcello-Sega @Yuan-Yu I've just added short pages on the new readers/parsers to the MDAnalysis user guide. These are mostly to highlight usage notes and unexpected behaviour, or point out features that are important to know, e.g. the TPR format starts atom ids from 0 despite other formats not doing this. It would be great to get your thoughts on if I should add or change anything here; I'm not familiar with any of the formats.

[Luthaf]

Looks good to me, I just find the 'chemfiles object' appellation a bit strange. What do you think of chemfiles' readers or chemfiles library readers or something like this?

[Marcello-Sega]

Thanks @lilyminium for requesting the feedback! I have only two minor comments

[Yuan-Yu]

Looks good to me for my part. Thank you for requesting the feedback. :)

[lilyminium]

Thanks @Yuan-Yu!

[lilyminium]

@orbeckst unless you have any changes to suggest for fnmatch and the new dependencies, I'll probably merge this in the next couple days :-)

[orbeckst]

lgtm – looking forward to the updated version