modularize optimization internals

[ben-schwen]

• Closes Using Map instead of lapply turns GForce off #5336
• Closes lapply GForce opt could work also without .SD #5032
• Closes Move GForce tests to own script #4305
• Towards GForce optimisation could be more smart #3815
• Closes GForce as.double / as.numeric #2934
• tests (a lot of them)

Adds arithmetic for GForce as demanded in #3815 but does not add support for blocks in j like d[, j={x<-x; .(min(x))}, by=y].

[codecov]

Codecov Report

❌ Patch coverage is 99.61089% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 99.11%. Comparing base (df7fa80) to head (c07999a).

Files with missing lines	Patch %	Lines
R/test.data.table.R	94.73%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #7401      +/-   ##
==========================================
- Coverage   99.13%   99.11%   -0.02%     
==========================================
  Files          85       85              
  Lines       16618    16701      +83     
==========================================
+ Hits        16474    16554      +80     
- Misses        144      147       +3     

☔ 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.

[github-actions]

• HEAD=modular_gforce slower P<0.001 for memrecycle regression fixed in #5463
  

Generated via commit c07999a

Download link for the artifact containing the test results: ↓ atime-results.zip

Task	Duration
R setup and installing dependencies	7 minutes and 27 seconds
Installing different package versions	44 seconds
Running and plotting the test cases	2 minutes and 40 seconds

[ben-schwen]

I'm also not sure about moving the tests to optimize.Rraw since this feels kind of wrong and not needed after introducing the new levels/optimization parameter to test.