ref: full state refactor (#4174)

## Current State Observations
- `Session` currently holds many unrelated responsibilities (history,
approval queues, task handles, rollout recorder, shell discovery, token
tracking, etc.), making it hard to reason about ownership and lifetimes.
- The anonymous `State` struct inside `codex.rs` mixes session-long data
with turn-scoped queues and approval bookkeeping.
- Turn execution (`run_task`) relies on ad-hoc local variables that
should conceptually belong to a per-turn state object.
- External modules (`codex::compact`, tests) frequently poke the raw
`Session.state` mutex, which couples them to implementation details.
- Interrupts, approvals, and rollout persistence all have bespoke
cleanup paths, contributing to subtle bugs when a turn is aborted
mid-flight.

## Desired End State
- Keep a slim `Session` object that acts as the orchestrator and façade.
It should expose a focused API (submit, approvals, interrupts, event
emission) without storing unrelated fields directly.
- Introduce a `state` module that encapsulates all mutable data
structures:
- `SessionState`: session-persistent data (history, approved commands,
token/rate-limit info, maybe user preferences).
- `ActiveTurn`: metadata for the currently running turn (sub-id, task
kind, abort handle) and an `Arc<TurnState>`.
- `TurnState`: all turn-scoped pieces (pending inputs, approval waiters,
diff tracker, review history, auto-compact flags, last agent message,
outstanding tool call bookkeeping).
- Group long-lived helpers/managers into a dedicated `SessionServices`
struct so `Session` does not accumulate "random" fields.
- Provide clear, lock-safe APIs so other modules never touch raw
mutexes.
- Ensure every turn creates/drops a `TurnState` and that
interrupts/finishes delegate cleanup to it.

codex-rs/core/src/state/turn.rs

@@ -0,0 +1,60 @@
+//! Turn-scoped state and active turn metadata scaffolding.
+
+use std::collections::HashMap;
+use std::sync::Arc;
+use tokio::sync::Mutex;
+
+use codex_protocol::models::ResponseInputItem;
+use tokio::sync::oneshot;
+
+use crate::protocol::ReviewDecision;
+
+/// Metadata about the currently running turn.
+#[derive(Default)]
+pub(crate) struct ActiveTurn {
+    pub(crate) sub_id: String,
+    pub(crate) turn_state: Arc<Mutex<TurnState>>,
+}
+
+/// Mutable state for a single turn.
+#[derive(Default)]
+pub(crate) struct TurnState {
+    pending_approvals: HashMap<String, oneshot::Sender<ReviewDecision>>,
+    pending_input: Vec<ResponseInputItem>,
+}
+
+impl TurnState {
+    pub(crate) fn insert_pending_approval(
+        &mut self,
+        key: String,
+        tx: oneshot::Sender<ReviewDecision>,
+    ) -> Option<oneshot::Sender<ReviewDecision>> {
+        self.pending_approvals.insert(key, tx)
+    }
+
+    pub(crate) fn remove_pending_approval(
+        &mut self,
+        key: &str,
+    ) -> Option<oneshot::Sender<ReviewDecision>> {
+        self.pending_approvals.remove(key)
+    }
+
+    pub(crate) fn clear_pending(&mut self) {
+        self.pending_approvals.clear();
+        self.pending_input.clear();
+    }
+
+    pub(crate) fn push_pending_input(&mut self, input: ResponseInputItem) {
+        self.pending_input.push(input);
+    }
+
+    pub(crate) fn take_pending_input(&mut self) -> Vec<ResponseInputItem> {
+        if self.pending_input.is_empty() {
+            Vec::with_capacity(0)
+        } else {
+            let mut ret = Vec::new();
+            std::mem::swap(&mut ret, &mut self.pending_input);
+            ret
+        }
+    }
+}