Issue 1601 match partial page citations

[mattdahl]

This is a first pass at addressing #1601, the other half of freelawproject/eyecite#30.

I've broken up our search_db_for_full_citation() function into two paths: one for when we have full citation information (which I've kept essentially the same as it is currently, just reordered slightly) and one for when we're looking for a case with a missing page citation. In the latter scenario, I've made it only return a result if the cited case was published within 5 years of the citing case, to guard against false positives.

The logic behind this is based on @brianwc's comment (#1226 (comment)):

That style of citation is generally only used in slip opinions prior to the
volume being published. However, if you know the year of the opinion in
which you found such a citation, then I think we'd find that the years
covered by the cited volume are that very year, maybe +/- 1 year. So, if I
find such a citation in an opinion from 2016, then the volume of that
citation likely covers opinions from 2016 as well, maybe 2015-17.

However, unfortunately the U.S. reporter, for one, takes a preposterous amount of time to assign page numbers, so for SCOTUS cases at least I think we need to go beyond a 1-year search frame. According to https://www.supremecourt.gov/opinions/USReports.aspx, it seems that as of today the U.S. reporter has only finalized its volumes for up to the 2015 term. I put in the 5 year range before checking this, but maybe it needs to be even longer?? If this is only a problem with SCOTUS cases, maybe we should treat them differently than all others, but I don't know if other reporters also have this problem. Obviously, the longer we allow the date range to be, the more false positives there will be.

Incidentally, how do we get finalized citations into CL when they're assigned so late? E.g. this case does not have its U.S. citation (575 U.S. 348) in the system: https://www.courtlistener.com/opinion/2795278/rodriguez-v-united-states

---

99a3bd3 is unrelated to the substance of this PR, but I realized as I was working on this that my factory changes in #2183 caused it to not be testing what we wanted anymore, so I fixed it.

[mlissner]

Thanks for this, Matt. I'm going to ask @flooie to review this one, because he's in deep with opinions and this will be a good opportunity for him to look at the citations code some more. Bill, I'll put this on your backlog so you can judge the priority of it against your other work. Thank you as always.

Matt, sorry to keep you waiting! We've got a few really big things coming together right now, so it's busy around here. I know you're probably not sweating it, but we like to be faster than this!