UsageEvent emission missing on /v1/audio/* (#226 follow-up)

[moonming]

Follow-up from #226 MVP (PR #402 lands /v1/embeddings).

Problem

`/v1/audio/transcriptions` and `/v1/audio/translations` (OpenAI Whisper-shape audio endpoints) do not emit `UsageEvent` to `ProxyState.usage_sink` on success. Same budget-ledger / /logs gap as embeddings.

Audio endpoints are billed differently from token-based endpoints — typically by audio duration in seconds. The `UsageEvent` shape will need adapting:

• `prompt_tokens` doesn't map cleanly. OpenAI's audio response does not return a usage block today.
• A custom telemetry field for `audio_seconds` may be needed cp-api-side, OR audio_seconds can be derived from the audio file's duration on the DP and emitted as `prompt_tokens` (cp-api recomputes cost from pricing catalog anyway).

Acceptance criteria

• On a successful 200 from /v1/audio/transcriptions or /v1/audio/translations, exactly one `UsageEvent` arrives on the configured sink
• Token/duration mapping documented in the PR (likely needs cp-api coordination on the receiving side)
• `model_id`, `api_key_id`, `inbound_protocol = "openai"` per chat.rs convention

References

• Parent issue: Non-chat endpoints don't emit UsageEvents — cp-api's spend ledger misses spend from /v1/responses, /v1/embeddings, /v1/audio*, /v1/images*, /v1/rerank #226
• MVP PR: feat(embeddings): emit UsageEvent on /v1/embeddings 200 (#226 MVP) #402 (embeddings)
• OpenAI audio API spec: https://platform.openai.com/docs/api-reference/audio
• Existing pattern: `crates/aisix-proxy/src/chat.rs::emit_usage_event`

[coderabbitai]

🔗 Related PRs

#355 - fix(proxy): use UUID request IDs across handlers [merged]
#363 - fix(proxy): record messages stream telemetry [merged]
#378 - feat(proxy): log routing target attempts [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!