See https://travis-ci.org/isislovecruft/bridgedb/jobs/16798991#L1591, it
was erroring because it expected the default hashring name for every
hashring which was prepopulated.

This class currently serves very little purpose because I am still
nervous about making major changes to the IPBasedDistributor and
EmailBasedDistributor classes. However, I did add functionality for
naming a Distributor's hashrings after the distributor's class, so that
which hashring a bridge is being added to is much clearer in the logs.

 * ADD bridgedb.Dist.Distributor class.
 * CHANGE IPBasedDistributor and EmailBasedDistributor to inherit from
   Distributor.
 * ADD functionality for naming hashrings after their distributor.

 * FIXES several lines which are too long.
 * FIXES a couple whitespace problems.

 * FIXES several PEP8 problems due to lines being too long.
 * FIXES one cute PEP8 problem; a semicolon which was placed after a
   `break` in a `for` loop, probably by someone whose fingers are used
   to coding in C.
 * FIXES one PEP8 problem due to continuation of an `if` clause on the
   same line.

Several `if` clauses are split to the next line, and an unneeded
line-continued backslash is removed from the parameters for a call to
the logger.

These are all whitespace changes ― nothing else is touched.

The logging before was quite messy.

The sub hashrings are identified by the HMAC of a unique string. It
turns out that the "unique string" which is used in this case is
actually something which looks like the class.__repr__() for a
bridgedb.Bridges.BridgeHolder subclass. This is *probably* unique
because it contains memory addresses of the newly generated hmac_fn()s,
which are likely (though not necessarily) unique. This is a bit gross,
in my opinion, but it's not really a problem.

However, the log messages this function generated are a problem. They
contain memory addresses, which isn't the greatest idea, ignoring all
other reasons, simply because it's illegible.

 * REFACTOR logging statements in
   bridgedb.Bridges.FilteredBridgeSplitter.prepopulateRings() to remove
   logging of hmac_fn() memory addresses.
 * ADD a long comment detailing the "unique strings" which identify a
   sub hashring, and how they should probably be changed someday.

In bridgedb.Bridges.FilteredBridgeSplitter.prepopulateRings(), we return
False if we were unable to prepopulate bridges into the sub hashring for
whatever reason. In case some calling function wants to use the boolean
return value as a check, we should mirror this logic and return True if
prepopulation was successful.

 * CHANGE bridgedb.Bridges.parseStatusFile() to only log networkstatus
   descriptor fields if the parsers were actually able to parse
   them. Otherwise we waste a lot of log lines with blank values.

[ci skip]

It was still failing.