handoff: VSync/event-wedge fixes + iterate 2.A–2.BC research notes

Source changes (dormant parity infra, retained from iterate 2.AI/2.AO):
- xenia-kernel/exports.rs: nt_create_event manual_reset polarity +
  related event wiring
- xenia-gpu/mmio_region.rs: D1MODE_VBLANK_VLINE_STATUS hardcode parity

Also lands the audit-runs/ analysis notes (.md/.txt/.json digests) for the
iterate 2.x VSync/0x10e8/0x1004 wedge investigation. Raw trace dumps
(.jsonl/.gz/.csv/.stdout) and agent worktrees (.claude/) are gitignored as
regenerable local artifacts — see memory + HANDOFF for the running findings.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

audit-runs/phase-c22-rtl-enter-leave-control-flow/broad-impact.md

@@ -0,0 +1,30 @@
+# Phase C+22 broad-impact (2026-05-18)
+
+## Resolved
+- None (escalation, no engine change).
+
+## Advanced
+- None on main chain (C+21 baseline 104,607 preserved).
+- None on sister chains (11/32/3/41/16 all preserved).
+
+## Persisted
+- The scheduler-determinism class persists. Same root cause
+  C+20 escalated. Multiple downstream effects of the same
+  asymmetry now identified at idx 104,608 (wait.begin —
+  absorbed by C+21) and idx 104,610 (post-wait nested-Enter
+  branch — NOT absorbable per reading-error #23).
+
+## NEW
+- Reading-error class **#34**: cold-run determinism depends on
+  input path form. Running ours against the loose `default.xex`
+  rather than the parent `.iso` produces a different boot
+  trajectory (40× more imports, 1.6M unimpl warnings, different
+  thread-create sequence). All cold-vs-cold protocol runs MUST
+  use the `.iso` path. ✓ documented in `investigation.md` and
+  `cold-vs-cold-result.md`.
+
+## Permanent infrastructure contribution
+- None (no engine, diff-tool, schema, or emitter changes).
+
+## Tests
+- 204 (unchanged from C+19/C+21).