return last company information as output file, not as warning

[jacobvjk]

closes #74

• instead of warning the users about companies lost in the sector_split, outputs are now generated that document the lost companies for later inspection
• data dictionary is updated accordingly with information about the new output file

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 19 lines in your changes missing coverage. Please review.

Project coverage is 51.20%. Comparing base (47f0663) to head (aaf4a9b).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
R/run_match_prioritize.R	0.00%	19 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #240      +/-   ##
==========================================
- Coverage   51.21%   51.20%   -0.02%     
==========================================
  Files          28       28              
  Lines        3118     3117       -1     
==========================================
- Hits         1597     1596       -1     
  Misses       1521     1521              

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

[cjyetman]

Makes sense and works as expected as far as I can tell locally. My only minor concern is that rather than spitting out a warning to the console for every loanbook included in the process (which could be 1000's maybe), now it will create a new CSV in the prioritized_loanbooks_and_diagnostics output directory for every loanbook included in the process? Maybe in the future we can think about how to better organize the prioritized_loanbooks_and_diagnostics output directory.

[jacobvjk]

Makes sense and works as expected as far as I can tell locally. My only minor concern is that rather than spitting out a warning to the console for every loanbook included in the process (which could be 1000's maybe), now it will create a new CSV in the prioritized_loanbooks_and_diagnostics output directory for every loanbook included in the process? Maybe in the future we can think about how to better organize the prioritized_loanbooks_and_diagnostics output directory.

Yes, that is a fair concern, structurally. For the moment I think this is fine, because the structure of the outputs will also be one file for each prioritized loan book, so there is some correspondence there. Also, the cases we have seen so far have always been significiantly below a case number of thousands of loan books. But I agree that it could be better nonetheless to gather all this information and write it into one file per run.