ci(windows): run E2E on PM test environment and platform markers

.github/workflows/tests-os.yml

@@ -173,9 +173,9 @@ jobs:
     # `hermes serve`, `hermes gateway run/stop`, the tui_gateway stdio process, cron script
     # jobs, the terminal tool's Git Bash / PowerShell / cmd children and a ConPTY console,
     # all against the loopback fake provider (tests/fakes/fake_llm_provider.py). Separate
-    # from os-tests above: those are unit-level windows_only tests, these spawn process
+    # from os-tests above: those are unit-level Windows tests, these spawn process
     # trees per test. Every file there is also marked `integration`, so os-tests'
-    # `-m "windows_only and not integration"` never runs them a second time.
+    # `-m "platforms and not integration"` never runs them a second time.
     #
     # One pytest process per FILE, in parallel (same model as the Linux `e2e` job): the
     # files spawn their own trees and must not share interpreter state. No retries: a
@@ -191,71 +191,23 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
 
-      - name: Install uv
-        uses: astral-sh/setup-uv@fac544c07dec837d0ccb6301d7b5580bf5edae39 # 8.2.0
+      - name: Set up locked Python and test dependencies
+        uses: ./.github/actions/setup-pm
         with:
-          version: "0.9.28"
-          enable-cache: true
-          cache-dependency-glob: |
-            pyproject.toml
-            uv.lock
-
-      - name: Set up Python 3.11
-        uses: ./.github/actions/retry
-        with:
-          command: uv python install 3.11
-
-      - name: Install dependencies
-        # Same extras as os-tests: serve needs the web stack, the gateway the messaging
-        # imports, and the ConPTY case pywinpty (a core dependency on win32).
-        uses: ./.github/actions/retry
-        with:
-          command: uv sync --locked --python 3.11 --extra all --extra dev --extra anthropic --extra mistral --extra fal --extra modal --extra daytona --extra parallel-web
-
-      - name: Minimize uv cache
-        run: uv cache prune --ci
+          packages: git,ripgrep
+          extras: '["all", "telegram", "anthropic", "mistral", "fal", "modal", "daytona", "hindsight", "parallel-web"]'
+          test-environment: 'true'
+          prune-python-cache: true
 
       - name: Run Windows E2E suite
         shell: bash
         run: |
-          set -uo pipefail
-          logs="${RUNNER_TEMP}/e2e-windows"
-          mkdir -p "$logs"
-          files=$(find tests/e2e/core/windows -name 'test_*.py' | sort)
-          if [ -z "$files" ]; then
-            echo "::error::no test files under tests/e2e/core/windows"
-            exit 1
-          fi
-          pids=()
-          names=()
-          start=$SECONDS
-          for f in $files; do
-            # `-m windows_only` replaces pyproject's `not integration` addopts selector.
-            timeout 900 .venv/Scripts/python.exe -m pytest "$f" -m windows_only \
-              -p no:cacheprovider -v --tb=short -rA > "$logs/$(basename "$f" .py).log" 2>&1 &
-            pids+=($!)
-            names+=("$f")
-          done
-          fail=0
-          for i in "${!pids[@]}"; do
-            rc=0
-            wait "${pids[$i]}" || rc=$?
-            f="${names[$i]}"
-            echo "::group::$f (rc=$rc)"
-            cat "$logs/$(basename "$f" .py).log"
-            echo "::endgroup::"
-            if [ "$rc" -eq 5 ]; then
-              echo "::error::$f selected zero tests (marker renamed or selection broken)"
-              fail=1
-            elif [ "$rc" -ne 0 ]; then
-              echo "::error::$f failed (rc=$rc)"
-              grep -E '^(FAILED|ERROR) ' "$logs/$(basename "$f" .py).log" || true
-              fail=1
-            fi
-          done
-          echo "Windows E2E wall: $((SECONDS - start)) s"
-          exit "$fail"
+          scripts/run_tests.sh tests/e2e/core/windows/ -- \
+            -m "platforms and integration" -v --tb=short -rA
         env:
+          HERMES_TEST_WORKERS: '6'
+          HERMES_TEST_FILE_TIMEOUT: '900'
+          HERMES_TEST_FILE_RETRIES: '0'
           OPENROUTER_API_KEY: ""
           OPENAI_API_KEY: ""
           NOUS_API_KEY: ""