Placetype local using English fallback and non-latin (latin) names

[nvkelso]

:wave: I did some awesome work for the Pelias project and would love for everyone to have a look at it and provide feedback.

Here's the reason for this change :rocket:

Connected to WOF issue https://github.com/whosonfirst-data/whosonfirst-data/issues/2154, I'm fleshing out some more edge cases, namely 2 cases:

1. Sometimes the official language for a place is not in a latin character set, but WOF has (yet another) property that we can use to add a parenthetical so locales and non-locals can still read the value.
2. Sometimes there is no official language placetype translation, but there is an English preferred placetype. When that's available then backfill the exported property with the English value.
   • NOTE: For the ~50k coarse admin places (country, dependency, macroregion, region, macrocounty, county) there is nearly 100% fill rate for label:eng_x_preferred_placetype because of recent WOF data work.

Here's what actually got changed :clap:

The code to calculate getPlacetypeLocal Shapefile property is extended with several new cases.

Here's how others can test the changes :eyes:

Example features:

• Makkah, Saudi Arabia
  • lang:ara_x_preferred_placetype =مقاطعة
  • lang:ara_latn_x_preferred_placetype = muhafazah
  • lang:eng_x_preferred_placetype = region
  • Expected placelocal value of مقاطعة (muhafazah)
• Glasgow, Scotland, UK
  • label:eng_x_preferred_placetype = unitary district
  • wof:placetype_local = unitary district
  • Expected placelocal value of unitary district
• Madrid, province in Spain – 
  • label:spa_x_preferred_placetype = provincia
  • label:eng_x_preferred_placetype = province
  • wof:placetype_local = autonomous community
  • Expected placelocal value of provincia

[missinglink]

The test suite is failing, looks like a typo and an easy fix. You can run the tests locally using the command npm test.

[nvkelso]

All tests (including new tests) pass. Ready for final review and merge :)