feat: wire up global soil ID algorithm

[garobrik]

Description

Ideally this will require minimal changes to models or GraphQL output.

Remaining work

• Get perf acceptable
• Remove unnecessary global shapefile check
• Fix outstanding bugs
• revert US
• Set up integration testing for soil-id-algorithm and backend
• Figure out HTML content in translation strings/move to i18n files (Carissa planning to do this)
• Fix typos:
  • 3 of the values under fao90_name in the hwsd2_data table seem to have typos. there are entries for:
    • Luvsiols instead of Luvisols
    • Albic Luvsiols instead of Albic Luvisols
    • Calcicols instead of Calcisols
    • CHERNOZEMS is in all caps
  • Gypsisols appears in hwsd2_data but not in wrb_fao90_desc, so description lookups fail for that soil type
• Add Makefile commands for dumping/restoring DB
• Try making many concurrent soil ID queries and see what happens to staging

Maybe now, maybe future?

• Move remaining contents of Data folder into soil-id-db
• remove sg_list stuff?

[DerekCaelin]

Goal is performance (looks like we're OK, at ~5s),
left: test backend multiple queries at once.
taking the opportunity to do more optimization, because when else? additional day of effort.
Need to figure out how to test our large geospatial resource. Another day.