Fix Test timeout

[ChristopherChudzicki]

What are the relevant tickets?

Closes #98

Description (What does it do?)

This PR changes one test that was being problematic.

How can this be has this been tested?

1. I ran the original test in a .each() block, repeated it 100 times, on both CI and locally. See https://github.com/mitodl/mit-open/actions/runs/6301758705/job/17107497440?pr=100
   • locally: that test file took ~130 seconds
   • on CI: 4 ❌, 96 ✅
2. I removed an unnecessary findByRole query, and repeated (1), running 100 tests. See https://github.com/mitodl/mit-open/actions/runs/6301882578
   • locally: that test file took ~55 seconds
   • on CI: 100 ✅

So, the test is a lot faster now.

Additional Context

Testing Library's *byRole queiries are their recommended query because they provide a number of accessibility checks, but they apparently slower than related queries that do not make similar accessibility checks (e.g., *byText*). See for example:

• waitFor + getByRole causing severe delays testing-library/dom-testing-library#820 (comment)
• Performance issue with the byRole query causing timeout errors testing-library/dom-testing-library#698 (comment)

But the performance is getting better: The unreleased branch of jest-environment-jsdom uses JSDom v22 (we're on v20). Trying that locally brought 100 repetitions down to undo 40 seconds.

[codecov]

Codecov Report

Merging #100 (2ce8137) into main (089b5c3) will decrease coverage by 0.02%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##             main     #100      +/-   ##
==========================================
- Coverage   78.02%   78.01%   -0.02%     
==========================================
  Files         323      323              
  Lines       14357    14357              
  Branches     2320     2320              
==========================================
- Hits        11202    11200       -2     
- Misses       2862     2863       +1     
- Partials      293      294       +1     

see 4 files with indirect coverage changes

[ChristopherChudzicki]

The button is actually the same element, just it's text changes. So there's no need to re-query the DOM to find it.