0ad34794

By: ValentinHirschi <valentin.hirschi@gmail.com>

bound exact CFF cache capacity and correct macOS memory guarding

Keep each exact-CFF request's normalized capacity in the cache key, preserving
its immutable numerator assignment and preventing independent terms from
forming an unnecessarily large Cartesian degree envelope. Measure the owned
process tree using native macOS physical footprint so compression and swap do
not evade the 30 GB guard. Reserve the existing large CI fixtures, correct the
JUnit path and Nextest minimum, and remove one unused Clippy lifetime binder.
Include the shared graph fixture directory in isolated Nix test build sources
so compile-time include_str! inputs are available to all three owning packages.

This remains a validation candidate, not ALL_GREEN. The complete scalar matrix,
fresh final focused/full gates and matching remote CI must still finish.

User prompt:
Ok continue as planned now, trying to finish this damn scalar LU matrix, and only looking for improvement to the code if you can find relatively non-invasive cheap wins. The core objective is getting to ALL_GREEN commit and push merge-ready state of the corresponding PR. Then stop yourself and deem the goal achieved (if there does not seem to be any gross oversight w.r.t what is in it).

Related user instruction:
I think the memory watchdog runner is not counting ram properly change it to follow just that process tree.
Also when rerunning can you start from GL16 directly?

Timeline