fix(mobile): keep thread scroll bounds current after animations

[rynfar]

Adopts #9013. Size-only content changes did not emit a scroll event, so the
list's "is at end" signal went stale after an animation settled — the thread
could think it was pinned to the bottom when it was not, and vice versa. Upstream
adds a getIsAtEnd that derives the edge from current content size rather than
waiting for a scroll event.

Carve-out

#9013 rewrites patches/@legendapp__list@3.3.5.patch, which carries Pylon's own
one-line fix: state.pendingMaintainScrollAtEnd = shouldMaintainScrollAtEnd
instead of = false, so the flag is deferred rather than discarded. Verified it
survives at both the CJS and ESM sites in the rewritten patch (lines 357 and 904),
since a patch-file merge is the one place a lost line produces neither a type
error nor a failing test.

The lockfile conflicted on the patch content hash and was regenerated rather than
hand-merged.

Verification

@t3tools/mobile typecheck clean, lint clean, 175 thread tests passing.
node scripts/release-smoke.ts passes, which is the check that actually exercises
the patch pin.

Model: Claude Opus 5 in 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.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.6 KiB	13.5 KiB	−38 B (−0.3%)	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	6.6 KiB	6.6 KiB	−40 B (−0.6%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	57.2 KiB	57.1 KiB	−44 B (−0.1%)	66.4 KiB	✅
Codex	Live turn messages	10	9	−1 (−10.0%)	21	✅
Claude	Total thread wire	13.8 KiB	13.5 KiB	−214 B (−1.5%)	15.1 KiB	✅
Claude	Thread snapshot wire	6.9 KiB	6.9 KiB	−3 B (−0.0%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.8 KiB	6.6 KiB	−211 B (−3.0%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	59.5 KiB	58.0 KiB	−1.5 KiB (−2.5%)	66.4 KiB	✅
Claude	Live turn messages	11	9	−2 (−18.2%)	21	✅

Baseline: 9378c6e · PR result: 0a67e69 · 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: 109.5 KiB
• Claude decoded thread snapshot: 110.2 KiB

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