Add GGUF audio bridge support

[Godzilla675]

Summary

• expose audio-capable helpers on the existing VLM/media path
• extend llama-test-cli with audio encode/generation coverage
• unblock downstream gguf_lib and ToolNeuron GGUF audio support

[Copilot]

Pull request overview

This PR extends the existing VLM/media pathway to support audio inputs by adding C API helpers for audio encode/generation, exposing audio capability metadata, and expanding llama-test-cli to exercise the new audio path.

Changes:

• Add ggml_engine_audio type plus new VLM audio helpers (ggml_engine_vlm_generate_audio, ggml_engine_vlm_encode_audio) and a bitrate query (ggml_engine_vlm_audio_bitrate).
• Implement audio bridging in the VLM implementation by reusing the existing “file bytes → mtmd bitmap” media decoder.
• Extend llama-test-cli with optional --audio tests for audio encode + generation.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 4 comments.

File	Description
engine/llama-test-cli.cpp	Adds CLI arg parsing for --audio and introduces audio encode/generation tests for VLM.
engine/ggml-engine.h	Extends the public C API with audio structs and new VLM audio-related functions.
engine/ggml-engine-vlm.cpp	Implements the new VLM audio helper functions and audio bitrate query.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 65d3072e4e

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".