feat(audio): add 32-voice PCM mixing and native music sequencing

Add self-contained SoundBanks and MusicBanks with per-sample PCM16 and IMA ADPCM storage, materialized to resident PCM16. Integrate instrument envelopes, automation, shared DSP and host lifecycle controls while preserving staged real-time execution.
This commit is contained in:
2026-11-20 22:15:24 +00:00
parent a32f834b2b
commit 8510c5b00d
40 changed files with 4708 additions and 509 deletions
@@ -379,3 +379,7 @@ Native boundaries MUST respect the [Registry typed-value and ownership contract]
## Fn Execution Boundary
Typed Fn callbacks execute as ordinary guest calls on the single VM context. Boot remains physical FunctionId 0; preparation requires an empty capture environment and existing argument checks. FRAME_SYNC preserves context for next-PC resumption; successful last-root RET completes entry and leaves rooted results before threshold GC. No native callback registration, retention or reentry API is introduced; raw HeapRef/ValueSnapshot copies do not establish guest roots or authorize cross-session imports. Syscalls remain non-first-class.
The `audio` and `music` modules both require the AUDIO capability. Their staged
commands, numeric ranges, costs and audio-time semantics are defined in the
[audio peripheral](05-audio-peripheral.md) and [bank/ABI tables](05a-audio-bank-format.md).