Split monolithic workflow into modular components

Separate workflows for flexibility:
- build.yml: Build verification (standalone + reusable)
- runtime.yml: Container & runtime tests with container lifecycle
- inference.yml: Inference tests with optional container management
- full-pipeline.yml: Orchestrates all stages with LLM judge

Each workflow can be triggered independently for targeted testing,
or run the full pipeline for complete validation.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

.github/workflows/inference.yml

@@ -0,0 +1,78 @@
+name: Inference Tests
+
+on:
+  workflow_dispatch:  # Manual trigger
+    inputs:
+      use_existing_container:
+        description: 'Use existing running container'
+        required: false
+        default: 'false'
+        type: choice
+        options:
+          - 'true'
+          - 'false'
+  workflow_call:      # Called by other workflows
+    inputs:
+      use_existing_container:
+        description: 'Container is already running'
+        required: false
+        default: false
+        type: boolean
+    outputs:
+      result:
+        description: "Inference test result"
+        value: ${{ jobs.inference.outputs.result }}
+
+env:
+  OLLAMA_HOST: http://localhost:11434
+
+jobs:
+  inference:
+    name: Inference Tests
+    runs-on: self-hosted
+    outputs:
+      result: ${{ steps.inference-tests.outcome }}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Install test runner dependencies
+        run: cd tests && npm ci
+
+      - name: Start container (if needed)
+        if: ${{ inputs.use_existing_container != 'true' && inputs.use_existing_container != true }}
+        run: |
+          cd docker
+          docker compose down 2>/dev/null || true
+          docker compose up -d
+          sleep 10
+
+      - name: Run inference tests
+        id: inference-tests
+        run: |
+          cd tests
+          # Progress goes to stderr (visible), JSON results go to file
+          npm run --silent dev -- run --suite inference --no-llm --output json > /tmp/inference-results.json || true
+
+          echo "--- JSON Results ---"
+          cat /tmp/inference-results.json
+
+      - name: Upload inference results
+        uses: actions/upload-artifact@v4
+        if: always()
+        with:
+          name: inference-test-results
+          path: /tmp/inference-results.json
+
+      - name: Stop container (if we started it)
+        if: ${{ always() && inputs.use_existing_container != 'true' && inputs.use_existing_container != true }}
+        run: |
+          cd docker
+          docker compose down || true
+          echo "Container stopped"