Elden Ring is too slow

This game is primarily CPU bounded but it's in a weird way.
The GXWorker threads spin a decent amount of time in a forward_list. The primary thread also walks these lists.
They seemingly pingpong work and never achieve 100% core utilization. The Primary thread is typically around 80-90% loaded, while the GXWorker threads are 15-40% loaded depending on platform.

It really just smells funny.

No files or tests are named. Start by profiling Elden Ring under FEX, focusing on GXWorker threads, the primary thread, and their forward_list traversal and work handoff. Compare thread utilization across platforms and document a reproducible finding or performance improvement as the completion criterion.

Written by the indexing model from the issue text.