feat: Wave C - pi-native steer/follow-up delivery

Delivery now rides pi's OWN follow-up queue when a run is live, and a
carry-everything idle wake otherwise. One registry, two vehicles;
claimSteers and continuation coalescing are deleted (owner rulings,
2026-09-26: follow-ups only; one queue; the wake is the only appender).

THE LIVE RUN (vehicle 1): the chat-handler registers the streaming
turn's agent (registerLiveAgent, next to registerTurn); the pump hands
every eligible queued marker to agent.followUp() - pi polls that queue
when the run would otherwise stop, injects the markers
(message_start/message_end on the live wire), and CONTINUES the same
run. followUpMode "all" joins every handed marker in one continuation
turn. The question stopper runs before the follow-up poll, so an open
quiz is never overridden - a handed marker strands at settlement and
waits (today's pacing). probe-handoffs never take this path (the
handoff must START the teacher's run - a record-level branch).

SETTLEMENT RECONCILIATION: settleSteers(chatId, persisted) at onFinish
(after unregisterLiveAgent) resolves+flips entries the run delivered
and returns stranded ones to pending - the registry stays the truth;
the dead agent's queued copy is abandoned. This is what makes the
handoff safe: pi owns no retry for a stranded follow-up.

THE IDLE WAKE (vehicle 2): waitForIdle (unchanged gates) then ONE turn
carrying EVERY pending marker as its input appends - dequeue-at-fire
(the Wave B regression lesson: the wake run's own follow-up poll must
never see its own cargo; a 409 pushes the cargo back and retries).

CLIENT: applyWireEvent now upserts non-assistant messages by id, so
in-run injected markers render live instead of dropping until settle.

Tests: jobs 44/44 (live handoff, settlement stranding, handoff-waits,
one-wake-carries-all); chat-flow 23/23 (the double-append pin drives
the real pump through the new input); session-core 28/28 (custom-role
wire appends). Full suite 997/997, tsc/lint/build green.

Live smoke (PORT=3004): research job delivered mid-probe - parked
behind the dangling questions exactly per semantics, landed once on
the open gates, teacher folded the brief in and re-dispatched; the
second (good) result rendered the full 10-source brief card and the
lesson continued from it. No duplicate ids anywhere in the record.

Known gap flagged to owner (NOT from this change): the server accepts
malformed answer events via raw API (answers without the answers[]
shape) and AnsweredCard then crashes the page render - surfaced by the
smoke's scripted answers, repaired by hand in the scratch record.

components/session-core.ts

@@ -290,7 +290,22 @@ export class SessionModel {
 					role?: string;
 					id?: string;
 				};
-				if ((message as { role?: string }).role !== "assistant") return;
+				if ((message as { role?: string }).role !== "assistant") {
+					// Wave C: markers injected mid-run (pi's follow-up delivery)
+					// arrive as custom-role message_start/message_end events -
+					// UPSERT by id (markers always carry steer_<id> ids; id-less
+					// non-assistant events never occur - injected messages emit
+					// start+end only, no streaming partials). Without this the
+					// live stream dropped the marker until the settle sync.
+					const mid = message.id;
+					if (mid === undefined) return;
+					this.mutate(() => {
+						const idx = this.tail.findIndex((m) => m.id === mid);
+						this.tail =
+							idx === -1 ? [...this.tail, message as LhMessage] : this.tail.with(idx, message as LhMessage);
+					});
+					return;
+				}
 				// pi streams AssistantMessages with NO id — the id exists only
 				// once the server's persistTurn stamps the record (the live
 				// streaming bug, 2026-09-26: minting a fresh id per event made