docs: sync README with recent changes (TTS, model support, Claude caching)

- Model support: text/vision now OpenAI-compatible only (Claude/Gemini
  via their OpenAI-compatible endpoints); drop stale native-protocol copy
  and the anthropic/google rows from the *_PROVIDER table
- TTS: document StepFun (step-tts-2, paid, better quality) alongside
  Xiaomi MiMo (free) across zh/en/ja; update Vercel deploy envDescription
- Claude: note that direct Anthropic endpoints lack caching, recommend
  gateway for full experience
- Cloudflare: preserve existing one-click deploy (compat work in progress)

.env.example

@@ -35,10 +35,12 @@ TEXT_BASE_URL=https://api.deepseek.com/v1
 TEXT_API_KEY=sk-xxx
 TEXT_MODEL=deepseek-v4-flash
 # TEXT_PROVIDER: openai_compatible (default). This is the ONLY supported text
-# protocol. To use Claude or Gemini, leave TEXT_PROVIDER unset and point at
-# their OpenAI-compatible endpoints:
-#   Claude  → TEXT_BASE_URL=https://api.anthropic.com/v1  TEXT_MODEL=claude-sonnet-4-6
+# protocol. To use Gemini, leave TEXT_PROVIDER unset and point at its
+# OpenAI-compatible endpoint:
 #   Gemini  → TEXT_BASE_URL=https://generativelanguage.googleapis.com/v1beta/openai  TEXT_MODEL=gemini-3.5-flash
+# For Claude, a compatible gateway (e.g. LiteLLM) is recommended — Anthropic's
+# official endpoint offers an OpenAI-compatible layer but no caching, raising
+# cost and latency for this app's prefix-cache-driven design.
 # TEXT_PROVIDER=openai_compatible
 
 # ---- 2. Image generator (renders the scene background) -------------