What does this PR do?

Fix /goal auto-continue loop never firing on gateway platforms (Telegram, Discord).

The continuation event was enqueued into adapter._pending_messages after _run_agent() had already returned and drained pending events. The event sat orphaned until the user sent another message, so the auto-continue loop never ran — turns_used stayed at 0 for all active goals.

Why CLI works but Gateway doesn't: The CLI (cli.py) has a continuous polling loop in _pending_input, so a synthetic enqueued message gets picked up immediately. The Gateway has no equivalent — it relies on platform adapters to trigger processing, and _enqueue_fifo() does not trigger the adapter.

Before the fix: continuation event enqueued → orphaned in _pending_messages → never consumed → /goal loop stops after first turn.

After the fix: continuation event enqueued → popped from _pending_messages → fed to _start_session_processing() → next turn fires immediately.

Related Issue

Fixes #28649

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)

Changes Made

• gateway/run.py: _post_turn_goal_continuation() — changed return type from None to bool. Returns True when a continuation event was enqueued, False otherwise.
• gateway/run.py: _handle_message() — after the finally block releases _running_agents state, checks the _goal_continuation_enqueued flag. If True, pops the actual pending event from adapter._pending_messages and feeds it to adapter._start_session_processing(), following the same pattern as _drain_pending_after_session_command in base.py.

How to Test

1. Unit tests: python -m pytest tests/gateway/test_goal_verdict_send.py tests/gateway/test_goal_status_notice.py -q
2. Manual (gateway): Set up a Telegram/Discord gateway. Send /goal <multi-step task>. Observe that continuation turns fire automatically after each response instead of requiring a user message.

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform: Windows 11

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A
• I've updated tool descriptions/schemas if I changed tool behavior — or N/A

Verification Log

Environment: Windows 11, Python 3.14.3