ci(k8s): gate the Helm chart on every PR (refs #168)

`make k8s-lint` existed since the chart landed but nothing ran it, so the chart had
no automated coverage at all. A `k8s` job now runs it plus `make k8s-drift` on every
push and PR: no cluster, ~20s, and it catches the two failure modes the chart is
actually exposed to — a values typo that renders invalid YAML, and a change made to
one stack but not the other.

helm is installed as its pinned static binary (the URL the Talos runbook already
gives developers) rather than via a marketplace action: nothing extra to vet.

The `k8s` targets stay out of `make ci` on purpose — helm is optional for anyone not
deploying to Kubernetes — which is the one place local and CI now differ, noted in
docs/runbooks/ci.md.

.gitea/workflows/ci.yaml

@@ -41,6 +41,27 @@ jobs:
             nuget-${{ runner.os }}-
       - run: make lint
 
+  # The Helm chart's only automated gate: it renders and schema-checks the whole
+  # stack, and checks it still describes the same stack as the compose file
+  # (ADR-0033). No cluster involved — see docs/runbooks/kubernetes-talos.md.
+  k8s:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: https://github.com/actions/checkout@v4
+      # helm as its pinned static binary rather than a marketplace action: one URL,
+      # the same one the Talos runbook §0 gives a developer, and no third-party
+      # action to vet (CLAUDE.md §13). The drift check also needs `docker compose`,
+      # which the runner already has (see docs/runbooks/ci.md).
+      - name: Install helm
+        run: |
+          mkdir -p "$HOME/.local/bin"
+          curl -sSL https://get.helm.sh/helm-v3.16.4-linux-amd64.tar.gz \
+            | tar xz -O linux-amd64/helm > "$HOME/.local/bin/helm"
+          chmod +x "$HOME/.local/bin/helm"
+          echo "$HOME/.local/bin" >> "$GITHUB_PATH"
+      - run: make k8s-lint
+      - run: make k8s-drift
+
   build:
     runs-on: ubuntu-latest
     steps: