refactor: switch the whole project to English

talk is going to be published. Code comments, CLI output, error messages,
daemon log lines, MCP descriptions and results, test names and assertions,
README, spec, skill, build scripts and flake are now English. Only the
spoken announcement texts stay bilingual: the wrap result and the test
sentence are German or English depending on the detected language.

Language fallback is now configurable: `lang = "en" | "de"` (env
TALK_LANG, default en) is used when detection cannot tell (numbers, single
words, ties) and for talk's own announcements. langdetect.Detect takes the
fallback as an argument instead of hard-coding German. `talk setup` asks for
it; `talk doctor` and `talk status` show it.

Vocabulary pinned across packages so the process-boundary tests observe
stable strings: chain log lines ("chain: tier X skipped (not available)",
"falling back to the next tier", "spoken via X"), queue lines ("discarded:
quiet hours ... active", "replaced by toast", "max_age"), CLI notes
("speaking directly", "daemon started at", "no daemon", "nothing to cancel"),
status block with aligned English labels, MCP messages, origin guard 403.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

internal/provider/doc.go

@@ -1,12 +1,11 @@
-// Package provider implementiert die TTS-Provider von talk und ihre
-// Fallback-Kette. Der Vertrag (Name/Available/Synthesize) trägt die
-// Vier-Stufen-Kette aus Task 03: ElevenLabs → OpenAI TTS → Edge-TTS →
-// OS built-in (überschreibbar per --provider/config provider). Die
-// Cloud-Provider liefern MP3 (Keys ausschließlich über TALK_ELEVENLABS_KEY
-// bzw. TALK_OPENAI_KEY, Endpunkte konfigurierbar und damit fake-bar);
-// osbuiltin liefert RIFF/WAVE-PCM. Die dokumentierte optionale Erweiterung
-// DirectSpeaker deckt Plattformen ab, deren OS-TTS keine Audio-Bytes
-// liefern kann (Linux speech-dispatcher) — Chain.Speak behandelt diesen
-// Fall inklusive tonlosem Verwerfen am null-Sink, damit CLI (Task 03) und
-// Daemon (Task 04) dieselbe Kette unverändert nutzen.
+// Package provider implements talk's TTS providers and their fallback
+// chain. The contract (Name/Available/Synthesize) carries the four-tier
+// chain: ElevenLabs → OpenAI TTS → Edge TTS → OS built-in (overridable via
+// --provider/config provider). The cloud providers return MP3 (keys come
+// exclusively from TALK_ELEVENLABS_KEY and TALK_OPENAI_KEY, endpoints are
+// configurable and therefore fakeable); osbuiltin returns RIFF/WAVE PCM.
+// The documented optional extension DirectSpeaker covers platforms whose
+// OS TTS cannot provide audio bytes (Linux speech-dispatcher); Chain.Speak
+// handles that case, including silent discarding on the null sink, so that
+// the CLI and the daemon use the same chain unchanged.
 package provider