feat(ci): isolate workflow evals into independent nightly job

- Splits 'Evals: Nightly' into 'evals' (general capabilities) and 'workflow-evals' (specific workflow simulations).
- 'workflow-evals' runs only on 'gemini-2.5-pro' (the target model).
- 'evals' excludes workflow tests to prevent noise/skewed metrics on other models.
- Removes code-level 'targetModels' restrictions in favor of CI configuration.
- Updates aggregation script to handle skipped tests correctly (though exclusion avoids them).

.github/workflows/evals-nightly.yml

@@ -70,7 +70,8 @@ jobs:
               $CMD -- -t "$PATTERN"
             fi
           else
-            $CMD
+            # Exclude workflow evals from the general matrix run
+            $CMD -- --exclude evals/workflows
           fi
 
       - name: 'Upload Logs'
@@ -81,9 +82,48 @@ jobs:
           path: 'evals/logs'
           retention-days: 7
 
+  workflow-evals:
+    name: 'Evals (Workflow) nightly run'
+    runs-on: 'gemini-cli-ubuntu-16-core'
+    steps:
+      - name: 'Checkout'
+        uses: 'actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8' # ratchet:actions/checkout@v5
+
+      - name: 'Set up Node.js'
+        uses: 'actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020' # ratchet:actions/setup-node@v4
+        with:
+          node-version-file: '.nvmrc'
+          cache: 'npm'
+
+      - name: 'Install dependencies'
+        run: 'npm ci'
+
+      - name: 'Build project'
+        run: 'npm run build'
+
+      - name: 'Create logs directory'
+        run: 'mkdir -p evals/logs'
+
+      - name: 'Run Workflow Evals'
+        continue-on-error: true
+        env:
+          GEMINI_API_KEY: '${{ secrets.GEMINI_API_KEY }}'
+          GEMINI_MODEL: 'gemini-2.5-pro'
+          RUN_EVALS: 'true'
+        # Explicitly target the workflow directory
+        run: 'npm run test:all_evals -- evals/workflows'
+
+      - name: 'Upload Logs'
+        if: 'always()'
+        uses: 'actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02' # ratchet:actions/upload-artifact@v4
+        with:
+          name: 'eval-logs-workflows-gemini-2.5-pro'
+          path: 'evals/logs'
+          retention-days: 7
+
   aggregate-results:
     name: 'Aggregate Results'
-    needs: ['evals']
+    needs: ['evals', 'workflow-evals']
     if: 'always()'
     runs-on: 'gemini-cli-ubuntu-16-core'
     steps: