Fix playAudio for Response streams

[fallintoplace]

Summary

• Convert Response Web ReadableStream bodies with Readable.fromWeb() before piping them to ffplay.
• Add a regression test using a native Response and mocked ffplay stdin.

Why

playAudio(response) treated Response.body as a Node stream and called .pipe() directly. Native fetch responses expose a Web ReadableStream, so .pipe() is not available and playback fails at runtime.

Validation

• ./node_modules/.bin/jest tests/helpers/audio.test.ts --runInBand
• ./node_modules/.bin/prettier --check src/helpers/audio.ts tests/helpers/audio.test.ts
• ./node_modules/.bin/eslint src/helpers/audio.ts tests/helpers/audio.test.ts
• ./node_modules/typescript/bin/tsc
• ./scripts/build

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: fb2dc04427

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[HAYDEN-OAI]

hey @fallintoplace, thanks so much for the fix and for your patience while we worked through the stream compatibility details! we’ve got both the web and node stream paths covered now, including clean error handling. merging this now 🎉