GitLab

* Fix embeddings memory corruption

The patch was leading to a buffer overrun corruption.  Once removed though, parallism
in server.cpp lead to hitting an assert due to slot/seq IDs being >= token count.  To
work around this, only use slot 0 for embeddings.

* Fix embed integration test assumption

The token eval count has changed with recent llama.cpp bumps (0.3.5+)

This reverts commit bb46bbcf.

* Fix assert on small embedding inputs

* Update llm/patches/09-pooling.diff

On windows, if the model dir contained unicode characters
clip models would fail to load.  This fixes the file name
handling in clip.cpp to support utf16 on windows.

* llm: update llama.cpp submodule to `7c26775`

* disable `LLAMA_BLAS` for now

* `-DLLAMA_OPENMP=off`

* update llama.cpp submodule to `5921b8f089d3b7bda86aac5a66825df6a6c10603`

* add patch

This doesn't expose a UX yet, but wires the initial server portion
of progress reporting during load

* fix llava models not working after first request

* individual requests only for llava models

The release just before ggml-cuda.cu refactoring

This should resolve the problem where we don't fully unload from the GPU
when we go idle.