Cap WASM parallel workers under a heap budget, Emscripten-only.

[tameware]

Do not request more than 2GB RAM from WASM, its limit. The messages generated upon exceeding it are apparently harmless, but best to avoid them.

$ bazel run //wasm:run_dtest_wasm -- -f hands/list1000.txt -s calc

…

Cannot enlarge memory, requested 2150969344 bytes, but the limit is 2147483648 bytes!

…

Keep the clamp behind EMSCRIPTEN so native resolve_worker_count stays uncapped.

[Copilot]

Pull request overview

This PR adds an Emscripten-only clamp to resolve_worker_count() so WASM batch runs don’t request more heap than wasm32 can provide (avoiding the “Cannot enlarge memory… limit is 2GB” message), while keeping native builds uncapped.

Changes:

• Introduce clamp_workers_to_memory_budget() to bound a requested worker count by a MB heap budget.
• Apply a fixed WASM heap budget cap inside resolve_worker_count() only when __EMSCRIPTEN__ is defined.
• Add unit tests covering the clamp behavior and verifying that the WASM cap is applied only under Emscripten.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
library/tests/system/worker_count_test.cpp	Adds tests for the new clamp helper and validates WASM-only capping vs native behavior.
library/src/system/parallel_boards.hpp	Declares the new clamp_workers_to_memory_budget() helper.
library/src/system/parallel_boards.cpp	Implements the clamp helper and applies it to resolve_worker_count() under __EMSCRIPTEN__.

[zzcgumn]

Maybe we will get a request to do something similar on a raspberry pi.

[tameware]

Maybe we will get a request to do something similar on a raspberry pi.

We might! I added issue #279 to generalize this fix.