search for RIs by name, inchi, or inchikey in addition to CAS

[Aariq]

This PR mainly addresses #219. It adds the ability to search for retention indices from NIST using InChI, InChIkey, or compound name.

Bugs fixed:

• I discovered and fixed a bug where compounds that only had one entry would produce malformed output.

Dependencies added:

• polite. I switched nist_ri() to use if for web scraping and I quite like using it, but I'm happy to remove this dependency if you don't like it.
• rlang. I did a few things with rlang that I do not know how to do otherwise.
• tibble. I switched the output to a tibble rather than a data.frame.

I'm not sure why so much of the README changed. It looks like maybe it just didn't get re-built with some previous PRs.

PR task list:

• Update NEWS
• Add tests (if appropriate)
• Update documentation with devtools::document()
• Update README examples
• Check package passed

[codecov-io]

Codecov Report

Merging #232 into master will not change coverage by %.
The diff coverage is 0.00%.

@@          Coverage Diff           @@
##           master    #232   +/-   ##
======================================
  Coverage    0.00%   0.00%           
======================================
  Files          19      19           
  Lines        1611    1672   +61     
======================================
- Misses       1611    1672   +61     

Impacted Files	Coverage Δ
R/nist.R	0.00% <0.00%> (ø)
R/chemspider.R	0.00% <0.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 434cd60...b050864. Read the comment docs.

[Aariq]

You know, I just realized I changed the scrape delay, but never tested this on large numbers of queries. Let me do some testing and see if I don't get my IP banned before merging this PR

[Aariq]

Ok, fixed it up so it works with NAs better. New tests are all passing locally.

[stitam]

Thanks @Aariq, this is great! Most of my comments were on the same topic, which is coding stye when connecting to other packages. I am not sure which is the best way to do this, but I see that we are not consistent within the package. @import, @importFrom, package::function(), what do you think?

I know I added v0.6.0 to NEWS.md but maybe it would be better to return to incrementing the version number with each PR, 0.5.1, 0.5.2, etc? http://r-pkgs.had.co.nz/release.html.

[stitam]

I ran R CMD check locally and it succeeded.