Enable wasm workers to run under node

[sbc100]

This means we can run the tests without depending on the browser.

[sbc100]

Will wait for @juj for final approval

[sbc100]

Actually I'm going to land this now since it seems fairly straight forward, and I could use this for debugging some other PRs. If the onmessage part of the change is controversial at all we can address that in a followup.

[kleisauke]

Interesting, I remembered that more changes were needed to fix Node.js compatibility, see e.g. PR #18201 (I should have split that PR back then, sorry).

[sbc100]

This change is minimal.. just enough to get the first few tests passing. There will more changes needed I'm sure to get the rest of them running.

[dschuff]

This seems to be causing a failure under LSan: https://logs.chromium.org/logs/emscripten-releases/buildbucket/cr-buildbucket/8781449586792126465/+/u/Emscripten_testsuite__LSan_/stdout

Aborted(Assertion failed: stackSize > tlsSize, at: /emsdk/emscripten/system/lib/wasm_worker/library_wasm_worker.c,47,emscripten_create_wasm_worker)
/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:142
   throw ex;
   ^

RuntimeError: Aborted(Assertion failed: stackSize > tlsSize, at: /emsdk/emscripten/system/lib/wasm_worker/library_wasm_worker.c,47,emscripten_create_wasm_worker)
    at abort (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:631:10)
    at ___assert_fail (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:3913:2)
    at emscripten_create_wasm_worker (<anonymous>:wasm-function[212]:0x6b28)
    at emscripten_malloc_wasm_worker (<anonymous>:wasm-function[213]:0x6b92)
    at __original_main (<anonymous>:wasm-function[46]:0xf76)
    at main (<anonymous>:wasm-function[47]:0xfb1)
    at /b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:657:20
    at callMain (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4912:13)
    at doRun (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4951:21)
    at run (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4963:3)
    at runCaller (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4901:18)
    at removeRunDependency (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:618:4)
    at /b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:889:44
    at Object.loadWasmModuleToAllWorkers (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:3753:3)
    at receiveInstance (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:889:11)
    at receiveInstantiationResult (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:897:3)
Thrown at:
    at abort (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:631:10)
    at ___assert_fail (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:3913:2)
    at emscripten_create_wasm_worker (:1:27433)
    at emscripten_malloc_wasm_worker (:1:27539)
    at __original_main (:1:3959)
    at main (:1:4018)
    at /b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:657:20
    at callMain (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4912:13)
    at doRun (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4951:21)
    at run (/b/s/w/ir/x/t/emtest_ttbbjo15/emscripten_test_lsan_0bdudi02/malloc_wasm_worker.js:4963:3)
    ```

[dschuff]

Actually I guess the problem is not new, it's just that we weren't running this test outside of the browser before, right?

[sbc100]

Right, old bug newly visible I would guess