feat: add qwen3.6-plus and set as default model

[Jing-ze]

Summary

• Add qwen3.6-plus to known models with 1M context window and 64K max tokens (image supported)
• Change the default model from qwen3.5-plus to qwen3.6-plus across all install scripts, CLI, configs, shell scripts, docs, tests, and CI
• Retain qwen3.5-plus as an available model option (case branches, known-models list, config templates)

Test plan

• Go build passes (go build ./...)
• Go tests pass (go test ./...)
• Python tests (copaw/tests/test_bridge.py) — needs pytest environment
• Verify install script menu shows qwen3.6-plus as option 1
• Verify hiclaw apply worker --name test defaults to qwen3.6-plus
• Verify switching to qwen3.5-plus still works via model-switch skill

🤖 Generated with Claude Code

[github-actions]

📊 CI Metrics Report

Summary

Metric	Current	Baseline	Change
LLM Calls	176	203	-27 ↓ -13.3%
Input Tokens	4284168	6209174	-1925006 ↓ -31.0%
Output Tokens	43415	41800	+1615 ↑ +3.9%
Total Tokens	4327583	6250974	-1923391 ↓ -30.8%

By Role

Role	Metric	Current	Baseline	Change
🧠 Manager	LLM Calls	130	147	-17 ↓ -11.6%
	Input Tokens	3396035	4798282	-1402247 ↓ -29.2%
	Output Tokens	25741	26902	-1161 ↓ -4.3%
	Total Tokens	3421776	4825184	-1403408 ↓ -29.1%
🔧 Workers	LLM Calls	46	56	-10 ↓ -17.9%
	Input Tokens	888133	1410892	-522759 ↓ -37.1%
	Output Tokens	17674	14898	+2776 ↑ +18.6%
	Total Tokens	905807	1425790	-519983 ↓ -36.5%

Per-Test Breakdown

Test	Mgr Calls	Wkr Calls	Δ Calls	Mgr In	Wkr In	Mgr Out	Wkr Out	Δ Tokens	Trend
02-create-worker	10	0	-1 ↓ -9.1%	238222	0	1148	0	-24736 ↓ -9.4%	✅ improved
03-assign-task	24	4	-1 ↓ -3.4%	588566	77426	3118	401	-21088 ↓ -3.1%	✅ improved
04-human-intervene	25	10	-3 ↓ -7.9%	659232	173763	5944	889	-99751 ↓ -10.6%	✅ improved
05-heartbeat	7	6	-2 ↓ -13.3%	229957	121503	2011	747	-110377 ↓ -23.8%	✅ improved
06-multi-worker	29	9	-21 ↓ -35.6%	701193	203470	6645	12715	-1338084 ↓ -59.2%	✅ improved
14-git-collab	35	17	+1 ↑ +2.0%	978865	311971	6875	2922	-329355 ↓ -20.2%	⚠️ regressed

Trends

✅ 5 test(s) improved (fewer LLM calls)
⚠️ 1 test(s) regressed (more LLM calls)

---

Generated by HiClaw CI on 2026-04-03 04:12:01 UTC

---

📦 Download debug logs & test artifacts

[johnlanni]

coding plan还没有支持qwen3.6-plus