Add a third fuzzer for @/$ equivalency

[MichaelChirico]

Part of #2191 -- continuing #2819. Also progress on #2737.

This one caught a pretty big gap in our current linter logic, as noted in #2820. Those are addressed here.

I left the new include_s4_slots=FALSE by default since apparently TRUE is only needed in one place to pass the current fuzz suite.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.27%. Comparing base (30d1e8e) to head (c11dc39).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2821   +/-   ##
=======================================
  Coverage   99.27%   99.27%           
=======================================
  Files         128      128           
  Lines        7165     7188   +23     
=======================================
+ Hits         7113     7136   +23     
  Misses         52       52           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[AshesITR]

Since we have 4 branches anyway, we might as well make them parallel, WDYT?

if (!is.null(function_names)) {
  if (include_s4_slots) {
    res <- combine_nodesets(function_call_cache, s4_slot_cache)
  } else {
    res <- function_call_cache
  }
} else {
  if (include_s4_slots) {
    res <- combine_nodesets(
      function_call_cache[names(function_call_cache) %in% function_names],
      s4_slot_cache[names(s4_slot_cache) %in% function_names]
    )
  } else {
    res <- function_call_cache[names(function_call_cache) %in% function_names]
  }
}

[MichaelChirico]

Works for me! The only downside I saw was repeating the clutter-ish names(function_call_cache) %in% function_names, so I pulled that into a variable. I think that also saves us the "embarassment" of having to disable unnecessary_nesting_linter() which is not equipped to allow "highly parallel" if/else structures like that.