test: make buffer sizes 32bit-aware in test-internal-util-construct-sab

[Renegade334]

Fixes: #61025
Refs: #60497

The range checks for bound SharedArrayBuffer construction weren't compatible with the maximum array buffer size on 32-bit architectures.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.53%. Comparing base (79df8b2) to head (0b36af9).
⚠️ Report is 10 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #61026      +/-   ##
==========================================
- Coverage   88.55%   88.53%   -0.02%     
==========================================
  Files         703      703              
  Lines      208522   208522              
  Branches    40219    40205      -14     
==========================================
- Hits       184647   184609      -38     
- Misses      15905    15919      +14     
- Partials     7970     7994      +24     

see 41 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[aduh95]

Don't we want to keep Number.MAX_SAFE_INTEGER + 1 test case?

[Renegade334]

The intention behind Number.MAX_SAFE_INTEGER was that this is the maximum array buffer size on 64-bit builds, so that particular test value lives on as kMaxLength + 1.

[aduh95]

IMO it's still interesting to test with an "unsafe" integer

[Renegade334]

That's kinda covered by 2^64, but can add this if preferred

[aduh95]

If it had to be one or the other, I would pick Number.MAX_SAFE_INTEGER + 1 over 2 ** 64, that seems more interesting / intent revelling to me. I don't feel strongly either way, so non-blocking comment

[Renegade334]

The significance of 2^64 is that the size is passed via an int64_t intermediate, so has potential to overflow if not correctly clamped. It's a little academic, I can't imagine any of our internals will be allocating terabytes of shared memory 😀