feat: implement Agent.stop() in python sdk

[shuaixr]

Summary

Implements Agent.stop() method that performs a clean async shutdown of an agent instance:

• Marks agent as shutting down and transitions status to OFFLINE
• Stops heartbeat background worker
• Notifies AgentField control plane of graceful shutdown (best effort)
• Cleans up async execution resources, memory event clients, and connection managers
• Idempotent: repeated calls have no additional effect after the first

Closes: #355

Testing

• ./scripts/test-all.sh
• Additional verification (please describe):

Checklist

• I updated documentation where applicable.
• I added or updated tests (or none were needed).
• I updated CHANGELOG.md (or this change does not warrant a changelog entry).

Screenshots (if UI-related)

Related issues

[CLAassistant]

All committers have signed the CLA.

[github-actions]

Performance

SDK	Memory	Δ	Latency	Δ	Tests	Status
Python	7.9 KB	-13%	0.33 µs	-6%	✓	✓

✓ No regressions detected

[github-actions]

📊 Coverage gate

Thresholds from .coverage-gate.toml: per-surface ≥ 86%, aggregate ≥ 88%, max per-surface regression ≤ 1.0 pp, max aggregate regression ≤ 0.50 pp.

Surface	Current	Baseline	Δ
control-plane	87.30%	87.30%	→ +0.00 pp	🟡
sdk-go	90.70%	90.70%	→ +0.00 pp	🟢
sdk-python	93.63%	93.63%	↑ +0.00 pp	🟢
sdk-typescript	92.56%	92.56%	→ +0.00 pp	🟢
web-ui	90.02%	90.01%	↑ +0.01 pp	🟢
aggregate	89.02%	89.01%	↑ +0.01 pp	🟡

✅ Gate passed

No surface regressed past the allowed threshold and the aggregate stayed above the floor.

[github-actions]

📐 Patch coverage gate

Threshold: 80% on lines this PR touches vs origin/main (from .coverage-gate.toml:thresholds.min_patch).

Surface	Touched lines	Patch coverage	Status
control-plane	0	—	➖ no changes
sdk-go	0	—	➖ no changes
sdk-python	0	—	➖ no changes
sdk-typescript	0	—	➖ no changes
web-ui	0	—	➖ no changes

✅ Patch gate passed

Every surface whose lines were touched by this PR has patch coverage at or above the threshold.

[AbirAbbas]

Nice addition — the try/except isolation around each cleanup step is the right call, and the idempotency guard works. A couple things worth addressing:

[AbirAbbas]

This calls clear_current_agent() but doesn't clear Agent._current_agent (the class-level attr). The existing _clear_current() method right above does both — it deletes the class attr and calls clear_current_agent().

After stop(), Agent.get_current() will still return the stopped agent. Should probably just call self._clear_current() here instead.