Hear speech instead of loudness in mavwaked (V-487)

silero-vad replaces the energy threshold when -vad-model points at it.
Everything after the speech decision is the same state machine: the speech
hold, the silence hold, the length cap and the utterance buffer.

The model window is 512 samples and the capture frame is 480, so silero.go
re-chunks across frames. main.go claimed the two matched, which was true of
silero v4.

Stage two, the wake word, is not here. It needs a Russian keyword model that
does not exist yet.

.gitignore

@@ -70,3 +70,5 @@ coverage.out
 
 # root .env — MAVEN_AMBIENT_TOKEN and friends, same class as deploy/telegram.env
 .env
+# silero-vad, downloaded (see AGENTS.md)
+/models/vad/