🔧 ci: pin Deno to v2.8.3 to work around v2.9.0 JSR warm-cache regression

[cowboyd]

What does this PR do?

Pins deno-version to v2.8.3 in size-report.yml and verify.yaml (previously v2.x, which floats to v2.9.0).

Deno v2.9.0 has a regression denoland/deno#35529 where JSR files that have no manifest checksum (e.g. @ts-morph/common@0.27.0's dynamic require("./crypto"), which resolves to a nonexistent file) pass on cold download but fail integrity check on warm cache:

error: Integrity check failed in package.
  Specifier: https://jsr.io/@ts-morph/common/0.27.0/crypto
  Actual:   e3b0c442…  (sha256 of empty bytes)
  Expected: package-manifest-missing-checksum

The Size Report workflow builds twice on one runner (head, then base) sharing ~/.cache/deno. The head build downloads fresh and passes; the base build hits the warm cache and errors — example failing run. Every Size Report run since 2026-06-30 has failed this way. Verify hasn't been bitten yet because it only builds once per runner, but I pinned it too so the next non-trivial refactor doesn't reintroduce a double build.

The upstream fix has been merged but no v2.9.1 tag exists yet. A comment in each workflow says to restore v2.x once it ships.

Type of change

• Chore (dependencies, CI, tooling)

Checklist

• All tests pass (pnpm test) — CI verifies
• Files are formatted (pnpm format)
• I have added/updated tests for my changes (if applicable) — n/a
• I have added a changeset — n/a, CI-only change

AI-generated code disclosure

• This PR includes AI-generated code

[pkg-pr-new]

Open in StackBlitz

npm i https://pkg.pr.new/@bomb.sh/tty@105

commit: 35219e9

[codspeed-hq]

Merging this PR will degrade performance by 10.29%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

❌ 2 regressed benchmarks
✅ 8 untouched benchmarks
⏩ 18 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	pack complex layout	47.8 ms	53.4 ms	-10.54%
❌	WallTime	pack large list	63.4 ms	70.5 ms	-10.03%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing fix-cache-bug (35219e9) with main (9de7762)²}

Footnotes

1. 18 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on main (5e253a7) during the generation of this report, so 9de7762 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[jbolda]

(In our meeting discussing this and) we acknowledge that this change shouldn't effectively be a regression. Something is going on and being a little funky with the results. We don't have our permissions squared away yet to mark this as an "Ok regression" (and is it really?) so it feels reasonable to proceed with merging.

[dreyfus92]

thank you both ✌🏻