frank-primary + frank-large: export /metrics, raise primary log verbosity to 4

Both b11030 serve scripts never passed --metrics, so /metrics answered
501 and Prometheus could not scrape frank-large (silently down since
its deploy) or frank-primary. frank-primary also gets -lv 4 (VERBOSITY
env tunable, default 4) for the full per-buffer load breakdown.

scripts/serve-frank-primary.sh

@@ -56,6 +56,7 @@ fi
 : "${BATCH_SIZE:=8192}"
 : "${UBATCH_SIZE:=512}"
 : "${CACHE_RAM:=16384}"
+: "${VERBOSITY:=4}"
 
 SERVER_BIN="$RUNTIME_DIR/llama-server"
 MODEL_PARTS=(
@@ -99,6 +100,8 @@ ARGS=(
     --alias frank-primary
     --ctx-size "$CONTEXT_SIZE"
     --parallel "$PARALLEL_SLOTS"
+    -lv "$VERBOSITY"
+    --metrics
     --n-gpu-layers "$NGL"
     --tensor-split "$TENSOR_SPLIT"
     --jinja