add optional cleanup page

Author: soyr-redhat

docs/lab/07-cleanup.md

@@ -0,0 +1,63 @@
+# Optional — Cleanup
+
+## Goal
+Remove the workshop resources from your OpenShift project in case you want to experiment with something different in the future.
+
+## What gets deleted
+- Model serving deployment and service
+- External route
+- PVC containing the model
+- (Optional) Workbench
+
+## Cleanup commands
+
+### 1. Delete the model serving resources
+
+Delete the deployment and service:
+
+```bash
+oc delete deployment mnist-onnx -n <PROJECT_NAME>
+oc delete service mnist-onnx -n <PROJECT_NAME>
+```
+
+### 2. Delete the route
+
+```bash
+oc delete route mnist-onnx -n <PROJECT_NAME>
+```
+
+### 3. Delete the model storage PVC
+
+```bash
+oc delete pvc mnist-model-pvc -n <PROJECT_NAME>
+```
+
+### 4. (Optional) Delete the Workbench
+
+If you want to remove the Workbench as well:
+
+**Via the OpenShift AI dashboard:**
+
+1. Go to your Data Science Project
+2. Click the **Workbenches** tab
+3. Click the three-dot menu next to `mnist-workbench`
+4. Select **Delete workbench**
+
+**Via CLI:**
+
+```bash
+oc delete notebook mnist-workbench -n <PROJECT_NAME>
+```
+
+!!! warning "This deletes your work"
+    Deleting the Workbench will remove the Jupyter environment and any unsaved notebooks. If you want to keep the notebook, download `mnist_sequential.ipynb` from JupyterLab before deleting the Workbench.
+
+## Verify cleanup
+
+Check that resources are deleted:
+
+```bash
+oc get deployment,service,route,pvc -n <PROJECT_NAME>
+```
+
+You should no longer see `mnist-onnx` resources listed.

mkdocs.yml

@@ -37,6 +37,7 @@ nav:
       - Step 3 - Deploy with OVMS: lab/04-serve.md
       - Step 4 - Test Inference: lab/05-test.md
       - Optional - External Curl Test: lab/06-external-test.md
+      - Optional - Cleanup: lab/07-cleanup.md
   - Troubleshooting: troubleshooting.md
   - Reference:
       - Repo assets: reference/repo-assets.md