ggml : add option to not print stack on abort (ggml/1081)

* Add option to not print stack on abort

Add option/envvar to disable stack printing on abort.
Also link some unittests with Threads to fix link errors on
ubuntu/g++11.

* Update ggml/src/ggml.c

---------

Co-authored-by: Diego Devesa <slarengh@gmail.com>

embedding : enable --no-warmup option (ggml-org#11475)

This commit enables the `--no-warmup` option for the llama-embeddings.

The motivation for this change is to allow the user to disable the
warmup when running the the program.

llama: fix missing k_cache store for rwkv6qwen2 (ggml-org#11445)

Signed-off-by: Molly Sophia <mollysophia379@gmail.com>

HIP: Supress transformation warning in softmax.cu

loops with bounds not known at compile time can not be unrolled.
when ncols_template == 0, the bounds of the loop are not constexpr, thus llvm cant unroll the loops here.

HIP: Only call rocblas_initialize on rocblas versions with the multip…

…le instantation bug (ggml-org#11080)

This disables the workaround on rocblas fixed versions (>=4.0.0) to eliminate the runtime cost and unnecessary VRAM allocation of loading all tensile objects.

Add github protocol pulling and http:// (ggml-org#11465)

As pulling protocols to llama-run

Signed-off-by: Eric Curtin <ecurtin@redhat.com>

cmake : don't fail on `GGML_CPU=OFF` (ggml-org#11457)

SYCL : SOFTMAX F16 mask support and other fixes (ggml-org#11261)

Implemented ggml_sycl_op_soft_max() F16 src1(mask) support for which a pragma deprecation warning was added during ggml-org#5021.
To do this, had to decouple it from ggml_sycl_op_flatten which always considered src1 to be of fp32 type(many OP functions are dependent on it).

* SYCL: SOFTMAX F16 mask support and other fixes

* test-backend-ops: Add F16 mask test cases

Handle missing model in CLI parameters for llama-run (ggml-org#11399)

The HTTP client in llama-run only prints an error in case the download of
a resource failed. If the model name in the CLI parameter list is missing,
this causes the application to crash.
In order to prevent this, a check for the required model parameter has been
added and errors for resource downloads get propagated to the caller.

Signed-off-by: Michael Engel <mengel@redhat.com>

Add new hf protocol for ollama (ggml-org#11449)

https://huggingface.co/docs/hub/en/ollama

Signed-off-by: Eric Curtin <ecurtin@redhat.com>