Rationalize Archer listing's use in tests

[anders-schneider]

We have a sample listing defined in ui-components/__tests__/fixtures/archer.json. It is used in a number of tests:

• ui-components/__tests__/page_components/UnitTables.test.tsx
• ui-components/__tests__/helpers/occupancyFormatting.test.tsx
• I believe many others (those two are just the two that I came across)

In some of these tests, the test itself defines UnitSummarys, which are intended to summarize the units that are defined in archer.json. In e.g. UnitTables.test.tsx, that summary comes with a note that it was "copied from listings service output", but I can't tell how to reproduce that output.

The units defined in archer.json and the unit summaries defined in the test could easily diverge, causing tests to fail in non-obvious ways. I'd propose that we either generate the summaries automatically (instead of defining them in the test) or explicitly define everything (the Archer listing and its summary) in the test itself.