stacks/voxtral: add entrypoint: vllm serve — image has no default ENTRYPOINT/CMD

Second voxtral attempt got past the image pull (v0.18.0 published,
~3 min download) but container init failed:
  unable to start container process: error during container init:
  exec: "--model=mistralai/Voxtral-4B-TTS-2603": stat ...: no such file

vllm/vllm-omni:v0.18.0 has Entrypoint=null AND Cmd=null — there's no
default executable. The compose's `command:` array becomes the full
exec invocation, with --model=... interpreted as the binary name.

Standard vLLM serving CLI is `vllm serve <model> [flags]`. The
binary's at /usr/local/bin/vllm. Set entrypoint: ["vllm", "serve"]
and pass the model as a positional arg.

While we're here: HF cache was empty too (Voxtral 4B BF16 ~8 GB
download on first start) — vLLM auto-downloads from HF on model
load, so no separate pre-pull step needed.

stacks/voxtral/compose.yaml

@@ -33,12 +33,16 @@ services:
     volumes:
       - ${VOXTRAL_CACHE_DIR}:/root/.cache/huggingface
       - ${VOXTRAL_VOICES_DIR}:/voices:ro
-    # vLLM-Omni's serve command — model + dtype + port pinned.
+    # vllm/vllm-omni image has no default ENTRYPOINT or CMD — the
+    # container init expected --model=... as argv[0]. Set entrypoint
+    # to `vllm serve` (the standard CLI) and pass model as positional
+    # + tuning flags via command.
+    entrypoint: ["vllm", "serve"]
     command:
-      - --model=${VOXTRAL_MODEL:-mistralai/Voxtral-4B-TTS-2603}
-      - --port=8000
-      - --dtype=bfloat16
-      - --gpu-memory-utilization=${VOXTRAL_GPU_UTIL:-0.85}
+      - "${VOXTRAL_MODEL:-mistralai/Voxtral-4B-TTS-2603}"
+      - "--port=8000"
+      - "--dtype=bfloat16"
+      - "--gpu-memory-utilization=${VOXTRAL_GPU_UTIL:-0.85}"
     healthcheck:
       # vLLM-Omni exposes /health for liveness + /v1/models for readiness.
       # /health 200 means the server's listening; /v1/models 200 means