[BUG] CAGRA extend operation does not support float16 data type

[TomaszGrabowskiPawlak]

Bug Description

CAGRA extend operation does not support float16 data type, while other CAGRA operations already support it.

Related Issues/PRs

• [BUG] CAGRA fails to load float16 cache. #1419 - Initial request for float16 support in CAGRA
• Fix C/Python serialization for FP16. Add python tests #1429 - Added float16 support for CAGRA build/search operations

Problem

PR #1429 added float16 support to CAGRA build and search operations, but the extend operation was not updated. This creates an inconsistency where users can build and search indices with float16, but cannot extend them.

Solution

Possible solution: https://github.com/TomaszGrabowskiPawlak/cuvs/pull/1/files

[TomaszGrabowskiPawlak]

@lowener , would you be so kind to have a look at this issue ? I've noticed that cagra doesn't support extend with float16 type and vibe-coded a patch ( https://github.com/TomaszGrabowskiPawlak/cuvs/pull/1/files ) , which I am currently using in my pipelines. However, not being experienced with cpp and rapids, I don't think it's acceptable for me to create a pull request.

[lowener]

It looks good, I created a PR starting from your work, and added some tests #1620