Optimize Smoke Claude workflow token spend (Haiku model, tighter turn cap, narrower GitHub toolset) (#2065)

* Initial plan

* ci: optimize smoke-claude model turns and github toolset

Agent-Logs-Url: https://github.com/github/gh-aw-firewall/sessions/12320643-5e08-4b46-84d1-2c6428335c6f

* chore: revert unintended actions-lock changes

Agent-Logs-Url: https://github.com/github/gh-aw-firewall/sessions/12320643-5e08-4b46-84d1-2c6428335c6f

* ci: update smoke-claude lock with model turns and trimmed github tools

Agent-Logs-Url: https://github.com/github/gh-aw-firewall/sessions/74ba0807-1761-43a1-bfb1-ad7a853aa1bf

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>

Author: Copilot

.github/workflows/smoke-claude.lock.yml

@@ -15,7 +15,8 @@ permissions:
 name: Smoke Claude
 engine:
   id: claude
-  max-turns: 12
+  model: claude-haiku-4-5
+  max-turns: 8
 features:
   cli-proxy: true
 strict: true
@@ -26,7 +27,7 @@ network:
     - playwright
 tools:
   github:
-    toolsets: [repos, pull_requests]
+    toolsets: [pull_requests]
   playwright:
   bash:
     - "*"
@@ -106,4 +107,4 @@ post-steps:
 
 If all tests pass, add the label `smoke-claude` to the pull request.
 
-**If triggered by workflow_dispatch or schedule** (no PR context), use a noop safe output to report the test results summary instead. Do NOT attempt to add comments or labels when there is no pull request.
+**If triggered by workflow_dispatch or schedule** (no PR context), use a noop safe output to report the test results summary instead. Do NOT attempt to add comments or labels when there is no pull request.