Better integrate trident with linetools package

[chummels]

Right now, Trident uses its own internal text file with line information about common ionic transitions, or it allows users to specify their own additional line lists with similar text files. The default text file is located in trident/trident/data/line_lists/lines.txt and contains 213 lines with information about names, ions, rest wavelength, gamma, and oscillator strength. As @peeples points out in issue #24 , it would be nice to have better integration between Trident and linetools a python package that many CGM/IGM observers are using that contains up to date information about many different lines.

This will require a little bit of thought on how best to integrate, and how the user will specify which lines they desire in an output spectrum, since linetools contains a lot more lines than the 213 lines currently in Trident. This issue can be a place where we discuss the best path forward before coding it up into a PR.

[qobilidop]

I've thought about this a little bit. My proposal is to provide an interface to convert a linetools AbsLine to a Trident Line, and a linetools LineList to a Trident LineDatabase. I've implemented my proposal in #44 for anyone to review. As I said in the PR, I'm not an expert in linetools. So please be critical of my solution.

[chummels]

I think this PR would be a great improvement to the code to ensure that we're always using spectral data that is up to date and what the observers are using. It also would increase the size of our spectral line database substantially beyond the contents of lines.txt. My main concern is that any times there are changes in the line database it will break our answer tests. I will talk to the authors of linetools to see how often they anticipate this occurring.