handle nested terminal if/else in return_linter()

[MichaelChirico]

Closes #2354, Closes #2356

NEWS is missing, but I'll put it into the expanded item for #2321 (depending on which is merged first)

[codecov-commenter]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (2dcb091) 99.11% compared to head (eaa36bc) 99.09%.

❗ Current head eaa36bc differs from pull request most recent head 0d57bc7. Consider uploading reports for the commit 0d57bc7 to get more accurate results

Files	Patch %	Lines
R/return_linter.R	97.43%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2362      +/-   ##
==========================================
- Coverage   99.11%   99.09%   -0.02%     
==========================================
  Files         125      125              
  Lines        5650     5658       +8     
==========================================
+ Hits         5600     5607       +7     
- Misses         50       51       +1     

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

[MichaelChirico]

I was also a bit worried about performance -- would overhead to recursing in R push us to pursue this as pure XPath?

I am satisfied with the R-recursion approach -- timing is not meaningfully different (faster, if anything) & should be much more readable. I linted all base R packages as a stress test:

setwd("src/library")
f = list.files()
f = f[file.exists(file.path(f, "DESCRIPTION"))]
lint_all = \(pkgs, linters) for (pkg in pkgs) lint_package(pkg, linters = linters)

# 'old' = current main, 'new' = current this PR

system.time(lint_all(f, old_explicit))
#    user  system elapsed 
#  72.638   1.162  74.112 
system.time(lint_all(f, old_implicit))
#    user  system elapsed 
#  66.546   0.819  67.682 
system.time(lint_all(f, new_explicit))
#    user  system elapsed 
#  72.178   1.058  73.604 
system.time(lint_all(f, new_implicit))
#    user  system elapsed 
#  62.563   0.871  63.674