chore(mobile): remove unused markdown archive and Metro dependency

[rynfar]

Mobile still declares an npm-only override pointing at an unused native-markdown 0.5.0 archive and a redundant direct Metro runtime dependency. Remove both, delete the archive, and stop copying it into the release smoke fixture. The regenerated lockfile removes only the direct Metro importer and marks the retained transitive snapshot optional; no package versions change.

Adopts the complete cleanup from T3 Code bed0b4cdad41408bb09b445a0a8f8ea0e5b62d1b (upstream #13148). This resolves the historical verification hold recorded in cycle #736, tracked by #865. The deleted archive is byte-identical to upstream's old artifact. Pylon's actual pnpm/EAS resolver already uses registry native-markdown 0.5.8 with its current patch; the resolved package path, patch hash, and installed md4c source hash are identical before and after. Pylon's custom entry imports Expo's root registration and has no direct Metro runtime import.

Validation: frozen offline mobile dependency installs before and after; iOS and Android production bundles before and after; mobile and scripts typechecks with incremental caches disabled; release smoke against a temporary manifest fixture; scoped lint/format and diff checks. iOS and Android development bundles also pass. Bundling verifies JavaScript dependency resolution, not a native device build. No EAS build, release, or installation.

Review cursor and frozen upstream bounds are unchanged.

Model: GPT-6. Harness: Codex desktop. Workflow: review-t3-upstream skill.

---

^{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.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	14.0 KiB	14.0 KiB	+16 B (+0.1%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.3 KiB	7.3 KiB	−11 B (−0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.7 KiB	6.8 KiB	+27 B (+0.4%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	58.0 KiB	58.0 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	14.0 KiB	14.0 KiB	−9 B (−0.1%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.3 KiB	7.3 KiB	−16 B (−0.2%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.7 KiB	6.7 KiB	+7 B (+0.1%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	58.9 KiB	58.9 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	9	9	0 (0.0%)	21	✅

Baseline: b5a607e · PR result: 2204eab · 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: 115.7 KiB
• Claude decoded thread snapshot: 116.4 KiB

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