Calling initialize_runtime_module twice in the same process leads to data race conditions

Prerequisites

Please put an X between the brackets as you perform the following steps:

• Check that your issue is not already filed:
  https://github.com/leanprover/lean4/issues
• Reduce the issue to a minimal, self-contained, reproducible test case.
  Avoid dependencies to Mathlib or Batteries.
• Test your test case against the latest nightly release, for example on
  https://live.lean-lang.org/#project=lean-nightly
  (You can also use the settings there to switch to “Lean nightly”)

Description

If from a C program you call

int main() {
initialize_runtime_module();
finalize_runtime_module();
initialize_runtime_module();
finalize_runtime_module();
}

then this both leaks memory and results in a data race, as the uv event loop started by the initializer is not stopped, and then is subsequently started for a second time in a new thread.

https://github.com/leanprover/lean4/pull/7752 is a rough attempt at fixing this.

Context

[Broader context that the issue occurred in. If there was any prior discussion on the Lean Zulip, link it here as well.]

Steps to Reproduce

1. Run the above C code under thread sanitizer or similar

Expected behavior: No errors

Actual behavior: Data race on event_loop->loop

Versions

Fails in v4.17.0.

Additional Information

Impact

Add 👍 to issues you consider important. If others are impacted by this issue, please ask them to add 👍 to it.

Start with the initialize_runtime_module and finalize_runtime_module entry points and reproduce the C example under ThreadSanitizer. Inspect the uv event loop state involved in the reported event_loop->loop race, and review pull request 7752; done means repeated initialization and finalization produce no leak or race.