fix(gallery): correct gemma-4 model URIs returning 404

[mvanhorn]

Description

This PR fixes #9361.

The gemma-4-26b-a4b-it, gemma-4-e2b-it, and gemma-4-e4b-it gallery entries point at files that do not exist on the upstream HuggingFace repos, so installing any of them fails with a 404. Multiple users in #9361 report the same behavior.

Two issues per entry:

• The mmproj filename uses the f16 quantization suffix, but ggml-org publishes the mmproj projectors as bf16 in these repos.
• The e2b and e4b URIs hardcode lowercase e2b / e4b in the filename component. HuggingFace file paths are case-sensitive, and the real files use uppercase E2B / E4B.

Example of the current failing URL and the working one:

404: https://huggingface.co/ggml-org/gemma-4-26B-A4B-it-GGUF/resolve/main/mmproj-gemma-4-26B-A4B-it-f16.gguf
302: https://huggingface.co/ggml-org/gemma-4-26B-A4B-it-GGUF/resolve/main/mmproj-gemma-4-26B-A4B-it-bf16.gguf

Updated filename, uri, sha256, and the top-level mmproj and parameters.model references so every entry points at a real file and the declared hashes match the content.

Notes for Reviewers

• SHA256s were pulled from the x-linked-etag header per .agents/adding-gallery-models.md.
• The 26B Q4_K_M file entry was already correct and was not modified. Only its sibling mmproj changed.
• Each of the six updated URIs was verified to return HTTP 302 against HuggingFace.

Signed commits

• Yes, I signed my commits.