feat: air-gapped deployment guide, no implicit downloads

- Ship tiktoken's cl100k_base inside the package and build the encoding
  from it, so token counting never downloads anything.
- Default EMBEDDINGS_CACHE_DIR to <data home>/models instead of FastEmbed's
  temp dir, and read tokenizer.json and repo metadata from that cache, so
  a model downloads once and survives reboots.
- TTS_PROVIDER=none and STT_PROVIDER=none switch the speech features off:
  the endpoints return 404, audio files fail to ingest with a clear
  message, /api/config reports tts_available/stt_available, and the UI
  hides the Speak and microphone buttons.
- Drop the Google Fonts Roboto import from the web UI.
- prefetch-models fills the cache the app reads; verify-offline checks the
  packaged encoding.
- Docs: new Air-Gapped Deployment guide, settings and cache notes.

docsgpt/tts/tts_creator.py

@@ -1,7 +1,11 @@
+from typing import Optional
+
 from docsgpt.tts.google_tts import GoogleTTS
 from docsgpt.tts.elevenlabs import ElevenlabsTTS
 from docsgpt.tts.base import BaseTTS
 
+#: TTS_PROVIDER value that switches text-to-speech off.
+DISABLED = "none"
 
 
 class TTSCreator:
@@ -10,9 +14,14 @@ class TTSCreator:
         "elevenlabs": ElevenlabsTTS,
     }
 
+    @staticmethod
+    def is_enabled(tts_type: Optional[str]) -> bool:
+        """False when the provider is ``none`` or empty: text-to-speech is switched off."""
+        return (tts_type or "").strip().lower() not in ("", DISABLED)
+
     @classmethod
     def create_tts(cls, tts_type, *args, **kwargs)-> BaseTTS:
         tts_class = cls.tts_providers.get(tts_type.lower())
         if not tts_class:
             raise ValueError(f"No tts class found for type {tts_type}")
-        return tts_class(*args, **kwargs)
+        return tts_class(*args, **kwargs)