feat: upgrade MiniMax default model to M2.7

[octo-patch]

Summary

• Add MiniMax-M2.7 and MiniMax-M2.7-highspeed to the model selection list
• Set MiniMax-M2.7 as the new default model when selecting MiniMax
• Retain MiniMax-M2.5 as an available alternative

Changes

• Model configs: Added M2.7 and M2.7-highspeed entries to known-models.json, manager-openclaw.json.tmpl, and worker-openclaw.json.tmpl (both models list and aliases)
• Shell scripts: Updated context window case statements in 4 scripts to recognize M2.7 and M2.7-highspeed
• Install scripts: Updated both bash and PowerShell installers — model display text, KNOWN_MODELS list, and default selection
• Docs: Updated model references in windows-deploy.md (EN + ZH) and both SKILL.md files

Why

MiniMax-M2.7 is the latest flagship model with enhanced reasoning and coding capabilities.

Testing

• All JSON configs validated
• No existing MiniMax unit tests to update (system uses integration tests)
• Changes are config/data-only — no logic changes

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
1 out of 2 committers have signed the CLA.

✅ octo-patch
❌ PR Bot

---

PR Bot seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[github-actions]

📊 CI Metrics Report

ℹ️ No baseline available - This is the first run or baseline data was not found.

Summary

Metric	Value
LLM Calls	233
Input Tokens	6100787
Output Tokens	49208

By Role

Role	LLM Calls	Input Tokens	Output Tokens
🧠 Manager	162	5109646	35272
🔧 Workers	71	991141	13936

Per-Test Breakdown

Test	Mgr Calls	Wkr Calls	Mgr In	Wkr In	Mgr Out	Wkr Out
02-create-worker	33	0	735685	0	5957	0
03-assign-task	15	13	380814	197521	2455	2462
04-human-intervene	20	9	506574	142060	3201	1220
05-heartbeat	8	1	270502	0	1782	0
06-multi-worker	36	23	1349222	284204	9249	3750
14-git-collab	50	25	1866849	367356	12628	6504

---

Generated by HiClaw CI on 2026-03-18 10:35:07 UTC

[johnlanni]

百炼coding plan还没有m2.7，可以先在known models里加上

[octo-patch]

Thanks for the review @johnlanni! Updated: reverted the coding plan model selector back to MiniMax-M2.5 (since Bailian coding plan does not support M2.7 yet), while keeping MiniMax-M2.7 and MiniMax-M2.7-highspeed in the known models list so they can be used via manual model switching. Please take another look.

[johnlanni]

LGTM