Fix Codex transcript streaming and images

PLAN.md

@@ -227,6 +227,28 @@ shown as unavailable while the free-text escape remains. Permission choices
 are likewise reported per provider: Codex offers its read-only,
 workspace-write and full-access modes, while Claude keeps its own modes.
 
+Resuming passes `excludeTurns: true`: this app already owns and pages its
+common transcript, so asking app-server to hydrate the complete Codex history
+only sends the rollout a second time. That is especially costly for image tool
+results, whose protocol records carry base64 data. Live structured tool
+results are split at the driver boundary: text becomes tool output and each
+image is saved under the session and emitted as `Image`, never serialized into
+a transcript line. Images attached to a remote Codex session ride the stdio
+protocol as inline image input, since the server's local attachment path does
+not exist on that machine. `thread/tokenUsage/updated.last.inputTokens` is the
+measured context (cached input is already included), while `last.totalTokens`
+remains the turn's usage. If an older common transcript has no such event yet,
+the server seeds the same measurement from the last `token_count` in Codex's
+own rollout, including when that rollout is on an SSH setup.
+
+App-server assistant text comes only from its durable
+`item/agentMessage/delta` notifications; the full text on `item/completed` is
+always the consolidated copy and is ignored. This is decided from the dialect,
+not an in-memory set of ids: after a backend restart, the previous deltas can
+be behind the persisted stdout cursor while the completion is still ahead,
+and forgetting which ids streamed used to append the complete message after
+its already-recorded prefix.
+
 ### The llama driver
 
 One `llama-server` per session, started through the same `Transport` as any