Clamp auto-compact limit to context window (#11516)

- Clamp auto-compaction to the minimum of configured limit and 90% of
context window
- Add an e2e compact test for clamped behavior
- Update remote compact tests to account for earlier auto-compaction in
setup turns

codex-rs/protocol/src/openai_models.rs

@@ -238,7 +238,8 @@ pub struct ModelInfo {
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub context_window: Option<i64>,
     /// Token threshold for automatic compaction. When omitted, core derives it
-    /// from `context_window` (90%).
+    /// from `context_window` (90%). When provided, core clamps it to 90% of the
+    /// context window when available.
     #[serde(default, skip_serializing_if = "Option::is_none")]
     pub auto_compact_token_limit: Option<i64>,
     /// Percentage of the context window considered usable for inputs, after
@@ -256,10 +257,16 @@ pub struct ModelInfo {
 
 impl ModelInfo {
     pub fn auto_compact_token_limit(&self) -> Option<i64> {
-        self.auto_compact_token_limit.or_else(|| {
-            self.context_window
-                .map(|context_window| (context_window * 9) / 10)
-        })
+        let context_limit = self
+            .context_window
+            .map(|context_window| (context_window * 9) / 10);
+        let config_limit = self.auto_compact_token_limit;
+        if let Some(context_limit) = context_limit {
+            return Some(
+                config_limit.map_or(context_limit, |limit| std::cmp::min(limit, context_limit)),
+            );
+        }
+        config_limit
     }
 
     pub fn supports_personality(&self) -> bool {