(WIP) v0.4 release

[shwestrick]

Working on a v0.4 release which incorporates entanglement management (paper to appear at PLDI'23) as well as other improvements along the way. Details to follow...

Here are some potential TODO items for v0.4. These aren't essential, but having the list is helpful regardless. We can discuss which we want to include and which we can leave for a future release.

Small things:

• Do -disable-pass splitTypes1 -disable-pass splitTypes2 by default? IIRC, these passes haven't been updated for polymorphic CAS. (We should open an issue about this.)
• We should change the wording in this error message (link). It's very old, and (with new entanglement management techniques) is misleading.

Big-ish things:

• 172 fixed the Real.toString bug. We should also consider fixing similar bugs, e.g. Real.fromString, Int.toString, some of the Time.{to,from}X functions, etc., all of which are inherited from MLton but which are not thread-safe by default.

• 169 discusses some performance problems with CGC. The second problem in particular (where an ancestor heap is pushed into the CGC chain, causing a space explosion) seems like it could be an easy fix with big impact.

[shwestrick]

Collecting a list of what we've done so far:

• bugfix: various toString/fromString/fmt functions are now thread-safe. (Used to be racy in parallel.)
• bugfix: disabled splitTypes1/splitTypes2 by default. These passes don't support polymorphic primitive CAS.
• bugfix: exception handling during CGC scheduling
• bugfix: polymorphic CAS on Real32.real array