Answer budget within the context window; enforced reviews of completed todos

Context window
- Requests asked for the model's full max_output, but providers count
  input + output against the window: long chats with large-output models
  (DeepSeek V4.1 Flash, 384k) failed with "input too long" before automatic
  compaction (85 %) kicked in. max_tokens is now what the window leaves
  (estimated input = last usage + messages since), the chat is compacted when
  less than 32k would remain for an answer, and a context-overflow error
  compacts and retries once.
- ModelConfig.max_input (catalogue limit.input, editable in the provider
  settings) lowers the compaction threshold for models with an input limit.

Enforced reviews (modes with refine.rounds > 0, refine.after_todos default on)
- Every todo marked completed is reviewed against the work since the last
  review; with findings it is reopened and the feedback goes to the agent
  (max. rounds per todo). Passed reviews leave a visible note.
- Ending the turn with open todos sends the agent back to work while it makes
  progress (max. 10 reminders in a row without progress); such modes may take
  up to 2,000 steps per turn. Reaching the step limit is reported.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

core/src/config.rs

@@ -64,6 +64,9 @@ pub struct ModelConfig {
     pub context: Option<u64>,
     #[serde(default)]
     pub max_output: Option<u64>,
+    /// Input limit when the provider sets one below the context window (e.g. 272k of 400k).
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub max_input: Option<u64>,
     #[serde(default)]
     pub reasoning: bool,
     #[serde(default)]