refactor: consolidate `diagnostics` funcs into single file

[agilgur5]

Summary

• Move diagnostics funcs into a single file named diagnostics.ts
  • Follow-up to #396 (and #413)
  • This should be the last of such "consolidation" PRs, as these 3 files were the only ones where I thought the separation was more confusing than helpful, and where consolidation simplified things, IMO
• Add a unit test for convertDiagnostic
• Tiny bugfix \n -> formatHost.getNewLine()
  • This doesn't seem to have ever affected anyone; I'm guessing either Node or people's terminals auto-translated the newlines b/t OSes or something? So this is more of a semantic bugfix in that sense

Details

refactor:

• move IDiagnostic and convertDiagnostic from tscache to print-diagnostics, then rename it to diagnostics.ts

  • the diagnostic funcs being in tscache always felt like a strange place for them
  • especially when parse-tsconfig or print-diagnostics would import them from tscache, which sounded like an unrelated file

• may want to move diagnostics-format-host into this file as well, as it is only used in this file

  • leaving as is for now, limiting this change to a smaller one

test: add unit test for convertDiagnostic

• this was previously only covered in integration tests

  • since unit tests don't currently touch index or tscache, and convertDiagnostic was previously in tscache
  • another reason why consolidating these functions into one file made sense

  fix(diagnostics): use formatHost.getNewLine() instead of \n

• since new lines are host OS specific

[agilgur5]

• Tiny bugfix \n -> formatHost.getNewLine()
  • This doesn't seem to have ever affected anyone; I'm guessing either Node or people's terminals auto-translated the newlines b/t OSes or something? So this is more of a semantic bugfix in that sense

Well this was pretty ironic -- the unit test I added actually fails on Windows without this fix (and a test adjustment). Given that I'm on Mac, I literally didn't see this until it hit CI after I wrote up the PR. Turns out the test confirms this behavior!

[agilgur5]

@ezolenko friendly ping in case you just hadn't seen this PR (since you merged some others, but not this one)

I also think we should be good to release 0.34.0, mainly containing #406 as that's a big enough change as is. 0.33.0 has already been out a few weeks without issues, so don't think we need to worry about releasing two minors shortly after each other anymore.

There's a couple more things that I'm working on, but those will probably take some time (not spending as much time on rpt2 anymore since I've fixed the vast majority of issues! 🙂), so don't want to hold back releasing #406 because of that, since it's a big fix for what is probably the single most common issue in this repo. Also the stuff I'm working on is mostly bugfixes and optimizations for cache/watch mode (i.e. follow-ups to #386), which can go out whenever as patch releases. #228 and #86 are the only real leftover bugs, and fixes to those probably should be in a separate minor than #406 anyway, IMO. Those might also be a few weeks away too (or more or less -- don't have much of a timeline on those in general), as they're not high priority in my "OSS todo list".

[ezolenko]

@agilgur5 0.34.0 is out on npm now, pls check if release notes on github make sense. Great work, thanks!