Refactor Scala definitions containing deeply nested classes to mitigate problems when accessing them from Java

[hubertp]

TreeToIr translates from new AST to Enso's internal IR. The class has been written in Java, by design and on purpose (reasons omitted), but it had to make use of the existing class hierarchy defined in IR.

While Scala and Java can co-exist, calling the former from the latter is not pleasant when dealing with nested classes. And Scala code often makes use nested classes. This is easily demonstrable just by looking at the imports of TreeToIr:

i.e. we can't use the typical "dot" syntax for nested classes. This makes it apparently impossible to recognize for IDE's, as demonstrated in the screenshot.

There are possible solutions (most of them hard to enforce because out of our control):

1. Make sure Scala emits better InnerClasses attribute for deeply (2+ level) nested inner classes. Not sure if this is even possible given the rather complex interop logic (e.g. https://discord.com/channels/@me/959434339182067712/1116738522003550269). Will file a ticket in the Scala compiler.
2. Make javac understand that nested classes belonging to companion objects are called Foo$Bar$Baz and not Foo$Bar$$Baz. Also unlikely to happen because most of the implementation follow blindly the $ pattern.
3. Fix IDE - unlikely to happen in a timely manner, given that IntelliJ is closed source
4. Refactor IR to multiple classes.

Due to the lack of other alternatives, this ticket proposes to pursue option 4. This bug slows down developer productivity quite significantly.

[hubertp]

Same situation with Runtime.scala

[enso-bot]

Hubert Plociniczak reports a new STANDUP for yesterday (2023-08-28):

Progress: Started refactoring IR so that we have a more a shallow hierarchy of classes. Sadly IDE refactoring didn't give too much support in this case. Added more debugging info to #7624, will wait for feedback. Waiting on PR review for logging. Some planning discussions. It should be finished by 2023-08-30.

Next Day: Next day I will be working on the #7017 task. Continue with refactoring.

[enso-bot]

Hubert Plociniczak reports a new STANDUP for yesterday (2023-08-29):

Progress: More refactoring work. It should be finished by 2023-08-30.

Next Day: Next day I will be working on the #7017 task. Continue with refactoring.

[enso-bot]

Hubert Plociniczak reports a new STANDUP for yesterday (2023-08-30):

Progress: Finished refactoring while waiting on CI to stabilize. It should be finished by 2023-08-30.

Next Day: Next day I will be working on the #7017 task. Wait on CI to resolve its problems and merge PR. Come back to logging PR.

[enso-bot]

Hubert Plociniczak reports a new 🔴 DELAY for yesterday (2023-08-31):

Summary: There is 1 day delay in implementation of the Refactor Scala definitions containing deeply nested classes to mitigate problems when accessing them from Java (#7017) task.
It will cause 0 days delay for the delivery of this weekly plan.

Delay Cause: Extra day to finish refactoring all classes

[enso-bot]

Hubert Plociniczak reports a new STANDUP for yesterday (2023-08-31):

Progress: Took some extra time to make sure that everything works as expected and cleaned up remaining craft (and waiting on CI to be sane again). Started looking at the logging PR feedback. It should be finished by 2023-08-31.

Next Day: Next day I will be working on the #7253 task. Back to addressing logging feedback