Add safer TUI defaults for cyber models (#37055)

## What changed

- Propagate the optional `modelSpecialty` field through the model catalog and
  app-server `model/list` response.
- When a newly selected model has the `cyber` specialty, default an active TUI
  thread to workspace-write permissions with on-request approval. Use auto
  review when available, otherwise keep the user as reviewer, while respecting
  configured permission requirements.
- Show a notice when auto review is applied and strengthen the full-access
  warning for cyber models.
- Preserve explicitly selected permissions when only reasoning settings change.

## Testing

- Cover auto-review defaults, requirement-aware fallback behavior, permission
  preservation across reasoning changes, and the new warning notice.

GitOrigin-RevId: 8f25b0082d3051d9ff91f7355921236d0f82f758

codex-rs/core/src/tools/handlers/multi_agents_spec_tests.rs

@@ -14,6 +14,7 @@ fn model_preset(id: &str, show_in_picker: bool) -> ModelPreset {
         model: format!("{id}-model"),
         display_name: format!("{id} display"),
         description: format!("{id} description"),
+        model_specialty: None,
         default_reasoning_effort: ReasoningEffort::Medium,
         supported_reasoning_efforts: vec![ReasoningEffortPreset {
             effort: ReasoningEffort::Medium,