voxtral + kyutai-tts: fix wrong image tag / wrong endpoint paths; fish-s2: env-selectable model variant

Three fixes from the second-wave deploy attempts:

* voxtral: vllm/vllm-omni doesn't publish a `latest` tag — pull
  failed with "manifest unknown". Pinned VOXTRAL_VLLM_TAG to v0.18.0
  (released 2026-03-29, the day after the Voxtral 4B TTS release —
  first cut with Voxtral support).

* kyutai-tts: NillPointer wrapper exposes ONLY /health (root) and
  POST /v1/audio/speech. No /v1/models, no /v1/audio/voices —
  those return 404. Verified by /openapi.json against the live
  container. Compose healthcheck + playbook wait + verify steps
  all repointed at the actual paths. POST /v1/audio/speech is now
  smoke-tested with a RIFF WAV assertion (same pattern as fish-s2).

* fish-s2: added FISH_S2_MODEL env var so the model variant is
  swappable via .env without rebuilding. Both s2-pro (default) and
  s1-mini are pre-pulled into the bind-mount; LLAMA_CHECKPOINT_PATH
  + DECODER_CHECKPOINT_PATH now use ${FISH_S2_MODEL:-s2-pro}.
  s1-mini was originally gated on fishaudio's HF org (401), but
  niobures/OpenAudio-S1 mirrors the same files openly — pulled
  from there via a one-shot snapshot_download.

stacks/voxtral/.env.example

@@ -3,9 +3,11 @@
 
 # ── image pin ────────────────────────────────────────────────────────
 # vLLM-Omni image tag (Mistral's partner serving stack for Voxtral).
-# Use a specific version rather than `latest` — vLLM moves fast and
-# Voxtral has version-specific compatibility.
-VOXTRAL_VLLM_TAG=latest
+# Pin a specific version — vllm/vllm-omni does NOT publish `latest`;
+# `:latest` 404s with "manifest unknown". v0.18.0 was released
+# 2026-03-29, one day after the Voxtral 4B TTS release, and is the
+# first vLLM-Omni cut with Voxtral support.
+VOXTRAL_VLLM_TAG=v0.18.0
 
 # Voxtral model on Hugging Face. The 4B variant is the only released
 # checkpoint as of 2026-04. Default BF16 weights are ~8 GB.