ci: run workflows on GitHub-hosted runners

[rynfar]

What Changed

Every runs-on in .github/workflows/ moves from a Blacksmith tier to the matching GitHub-hosted standard runner: ubuntu-24.04, windows-2025, macos-26. The Android showcase emulator drops from 8 to 4 cores to fit the 4-vCPU hosted runner. The four comments that explained Blacksmith-specific behaviour are rewritten, the Runners sections in docs/operations/release.md and docs/operations/mobile-app-store-screenshots.md describe the hosted runners, and AGENTS.md describes the repository as public.

No Blacksmith-specific action (useblacksmith/*, sticky disks, cache) was in use, so the label swap is the whole migration. actionlint reports only three pre-existing findings at lines this branch does not touch.

Why

Blacksmith was the CI bill: ~60 CI runs over three days at 7 Linux jobs each, plus a daily nightly with two macOS legs and a Windows leg. Standard GitHub-hosted runners are free on public repositories, and the repository went public today, so hosted runners bring the CI cost to zero with no infrastructure to keep alive.

Trade-off: hosted Linux runners are 4 vCPU against Blacksmith's 8 and hosted macOS is ~3 vCPU, so jobs take longer. Timeouts are unchanged; this PR's own run is the first measurement.

Checklist

• This PR is small and focused
• I explained what changed and why
• I included before/after screenshots for any UI changes (no UI change)
• I included a video for animation/interaction changes (no motion change)

Model: Claude Fable 5 · Harness: Claude Code

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

ℹ️ The exact PR base did not have a successful artifact. Baseline uses the latest successful main measurement shown below.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.3 KiB	13.5 KiB	+188 B (+1.4%)	15.1 KiB	✅
Codex	Thread snapshot wire	6.9 KiB	6.9 KiB	+2 B (+0.0%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.6 KiB	+186 B (+2.8%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	55.6 KiB	57.0 KiB	+1.5 KiB (+2.6%)	66.4 KiB	✅
Codex	Live turn messages	10	11	+1 (+10.0%)	21	✅
Claude	Total thread wire	13.4 KiB	13.3 KiB	−42 B (−0.3%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	−6 B (−0.1%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−36 B (−0.5%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	56.4 KiB	56.4 KiB	−44 B (−0.1%)	66.4 KiB	✅
Claude	Live turn messages	11	10	−1 (−9.1%)	21	✅

Baseline: 3a8d127 · PR result: 124630c · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 109.4 KiB
• Claude decoded thread snapshot: 110.0 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.