fix(llm-serving): retire the reference clef-serve, free the GPU

clef-serve (pytorch reference) and clef-candle both claim replicas 1, but only
one GPU is free: coder takes 2 of the 4 V100s. They starved each other, so
clef-candle sat Pending and the clef-serve Service had no endpoints.

The Service now selects app: clef-candle, so the reference receives no traffic
even when it runs. Keeping it at 1 only burns a GPU.

k8s/apps/llm-serving/clef-candle.yaml

@@ -0,0 +1,92 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: clef-candle
+  namespace: llm-serving
+  labels:
+    app.kubernetes.io/part-of: llm-serving
+    app: clef-candle
+spec:
+  replicas: 1
+  strategy:
+    # One replica holds one GPU, so Recreate keeps the rollout to a single
+    # weights load instead of needing a second card for a surge pod.
+    type: Recreate
+  selector:
+    matchLabels:
+      app: clef-candle
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/part-of: llm-serving
+        # The clef-serve Service now selects this label (see clef-serve.yaml),
+        # so clients keep addressing clef-serve:80 unchanged. The reference
+        # pytorch Deployment must stay at replicas 0 while this runs, or the
+        # Service splits traffic across two different implementations.
+        app: clef-candle
+    spec:
+      # Required. Without the nvidia runtime class libcuda.so is never injected
+      # and candle-vllm panics in cudarc at startup.
+      runtimeClassName: nvidia
+      containers:
+        - name: candle
+          # :sm70 is built by .gitea/workflows/candle-vllm-sm70.yaml on every
+          # merge to main. It MUST be a build that contains the GDN conv fix
+          # (candle-vllm PR: feat/clef-systemone). Older :sm70 images zero the
+          # key channel band in causal_conv1d_fwd, which starves the delta rule
+          # and makes every linear-attention layer return zero. The symptom is
+          # a well-formed /v1/systemone response with wrong answers, not an error.
+          image: forgejo.riotpiao.com/riotpiao-poimen/candle-vllm:sm70
+          # :sm70 is a moving tag. Default IfNotPresent keeps a cached copy and a
+          # rebuilt image never reaches the pod, so a merged fix looks like it did
+          # not land. Always forces the pull on every pod start.
+          imagePullPolicy: Always
+          # Runs the fixed binary from the models PVC, not the one in the image.
+          # The published :sm70 predates the /v1/systemone route and the GDN conv
+          # fix (candle-vllm f3c00b2 is merged but CI has not republished the tag),
+          # so the in-image binary 404s on the route. The PVC copy at
+          # /mnt/models/bin/candle-vllm contains both fixes.
+          #
+          # RETIRE THIS once CI publishes a :sm70 built from f3c00b2: change the
+          # command back to /usr/local/bin/candle-vllm and delete the PVC copy.
+          command: ["/mnt/models/bin/candle-vllm"]
+          # clef-flash is a hybrid Qwen3_5 model: 8 full-attention layers plus 24
+          # GDN layers that carry per-sequence recurrent state. It runs on ONE
+          # card because the state is not shardable the way attention KV is.
+          args:
+            - --w
+            - /mnt/models/clef-flash
+            - --d
+            - "0"
+            - --dtype
+            - f16
+            - --p
+            - "8080"
+            - --max-num-seqs
+            - "4"
+            # About 66.8K KV tokens. Clef is a single-shot classifier-style
+            # endpoint, so it does not need the coder's pool.
+            - --kv-fraction
+            - "0.30"
+            # Prefix caching stays on: clef requests repeat the schema block.
+            - --prefill-chunk-size
+            - "1024"
+          ports:
+            - containerPort: 8080
+          resources:
+            limits:
+              cpu: "8"
+              memory: 32Gi
+              nvidia.com/gpu: "1"
+            requests:
+              cpu: "4"
+              memory: 16Gi
+              nvidia.com/gpu: "1"
+          volumeMounts:
+            - name: models
+              mountPath: /mnt/models
+              readOnly: true
+      volumes:
+        - name: models
+          persistentVolumeClaim:
+            claimName: llm-models

k8s/apps/llm-serving/clef-serve.yaml

@@ -43,7 +43,7 @@ spec:
   # Cold start is about four minutes: the pod pip-installs on every boot because the
   # torch base image cannot be pushed to the registry (Cloudflare answers 413 above
   # 100 MB per layer). startupProbe allows for that.
-  replicas: 1
+  replicas: 0
   # Recreate, not RollingUpdate. Every GPU card is allocated, so a rolling update leaves
   # the new pod Pending forever waiting for a card the old pod still holds.
   strategy:
@@ -127,8 +127,11 @@ metadata:
   labels:
     app.kubernetes.io/part-of: llm-serving
 spec:
+  # Points at the candle-vllm implementation (clef-candle.yaml). Switching back
+  # to the reference pytorch server is this line plus that Deployment's replicas;
+  # no client changes, because the Service name and ClusterIP do not move.
   selector:
-    app: clef-serve
+    app: clef-candle
   ports:
   - name: http
     port: 80

k8s/apps/llm-serving/kustomization.yaml

@@ -21,6 +21,10 @@ resources:
   # streams against 71.5 ms and 215 tok/s for the 4-bit MoE. The V100 has FP16
   # tensor cores and no INT4 ones, so 4-bit pays dequantisation on the SM while
   # the memory controller idles at 19%.
+  # clef-candle.yaml serves clef-flash on candle-vllm (hybrid Qwen3_5, one GPU)
+  # and owns the clef-candle label that clef-serve.yaml's Service now selects.
+  # It must not sync against an image built before the GDN conv fix.
+  - clef-candle.yaml
   - coder-14b-fp16.yaml
   - clef-serve.yaml
   # Stable client-facing name for the chat model. poimen-memory points at this Service,