Set Codex lane defaults to GPT-5.5 xhigh fast

[MagMueller]

Summary

• make new Codex lanes default to GPT-5.5, xhigh reasoning, and fast service tier on
• make /model reset restore those bux defaults
• keep /fast as a binary fast-service-tier toggle

Tests

• /home/bux/.local/bin/ruff check agent/telegram_bot.py agent/test_telegram_bot.py
• /opt/bux/venv/bin/python -m unittest agent/test_telegram_bot.py

---

Summary by cubic

Set Codex lane defaults to GPT-5.5 with xhigh reasoning and fast service tier. /model reset now restores these defaults; /fast remains a simple on/off toggle.

• New Features
  • New lanes and cleared settings default to model gpt-5.5, reasoning xhigh, service tier priority.
  • /fast toggles the service tier between "priority" and "off" and updates messages accordingly.
  • _codex_settings_for returns defaults instead of an empty dict; _set_codex_settings(clear=True) returns defaults.
  • Persist service tier explicitly as "priority" or "off"; invalid inputs are ignored and defaults preserved.
  • Removed "Default" buttons from the model/effort picker UI and updated tests to match the new behavior.

^{Written for commit a3b19e6. Summary will update on new commits. Review in cubic}

[cubic-dev-ai]

No issues found across 2 files

_{Re-trigger cubic}