[Prototype] Reuse one dispatch fiber and batch await continuations (follow-up to #6)

[jbaczuk-qualia]

Summary

Optional follow-up to #6 (stacked on it; merge #6 first). Needs qualialabs/node#9. Instead of a new hidden fiber per dispatched job, keep one long-lived hidden dispatch fiber, and after each job let it run the dispatchable jobs queued right behind it, so a chain of awaits in fibered code costs one switch onto the fiber and one back instead of two per await.

File	Change
fibers_async.js	reusingDispatcher: runs jobs on the dispatch fiber and, while it is still the dispatch fiber, keeps calling Fiber.__runNextDispatchable(). A job that blocks keeps the fiber (it's parked inside that job) and the next dispatch starts a new one, the same trick as znewsham's drain fiber in this repo's #7. A parked fiber that is later resumed never pulls in other jobs. A dispatch that arrives while the dispatch fiber itself is running (nested drain) falls back to a one-off fiber. FIBERS_AWAIT_REUSE=0 (or a node without node#9) gives the #6 behaviour; Fiber.microtasksBatched counts jobs run without a switch
src/fibers.cc	Fiber.__runNextDispatchable(), defined only when dlsym finds v8_qualia_RunNextDispatchableMicrotask

Testing

On #6's node 18 build plus node#9, arm64 dev-qualia container:

• My standalone checks: 11/11 with reuse on and off (same 45 jobs: 27 dispatched + 18 batched with reuse, 45 dispatched without).
• znewsham's test-microtask-fibers.js (with Fiber.current swapped for the running fiber, poolSize = 1e9): same as on [Prototype] Run await continuations from fibers in hidden fibers (needs node#7) #6, 7/7 behaviour tests pass including "a parked microtask does not stall other promises" and "resuming a parked fiber does not drain unrelated microtasks mid-callback"; the 100k park/resume cycle test still exceeds its 10 s timeout (~9.7 s for the cycles alone; every park needs its own fiber).
• meteor-promise suite: now 15/15 with native Promise too (the extra fiber that [Prototype] Run await continuations from fibers in hidden fibers (needs node#7) #6 caused in should not double-wrap callbacks is gone).
• test/: unchanged (pool.js/cleanup.js pthread coro_destroy segfault as on stock).
• Local qli5 qualia boots; startup hook that awaits then blocks works; fiberless code still throws Can't wait without a fiber.

	#6	#6 + this
chain of awaits in a fiber	~25–30 us	0.13 us (stock: 0.05–0.07 us)
I/O-style await in a fiber	~33 us	~28 us (stock: ~1.1 us; ~20 us is the pthread run+yield floor)
drains / awaits in code that never had a fiber	unchanged	unchanged

🤖 Generated with Claude Code