Force FP32 compute in GLM4 FFN Down (ggml-org#13101)

* Force FP32 compute in cuBLAS GEMM

* Revert "Force FP32 compute in cuBLAS GEMM"

This reverts commit 6efd872.

* Force F32 compute in GLM4 ffn down

* Edit comment to clarify issue

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>

---------

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>

clip : fix pixtral on some GPU backends (ggml-org#13097)

* clip : fix pixtral on some GPU backends

* refactor inp_raw set

* rm outdated comment

* fix dynamic size

* add TODO

change the reorder tensor from init to execute OP (ggml-org#13003)

rpc : do not wait for response when sending RPC_CMD_SET_TENSOR (ggml-…

…org#12943)

RPC_CMD_SET_TENSOR always returns an empty response and we send this 4
times per token. We can improve TG speed if we don't wait for this empty
response.

The performance impact of this change depends on the network latency.

clip : remove boi/eoi embeddings for GLM-edge model (ggml-org#13081)

embeddings : fix batch sizes (ggml-org#13076)

ggml-ci

ggml : fix trailing whitespaces (#0)

CUDA: use switch statements in constexpr functions (ggml-org#13095)

cmake : do not include ./src as public for libllama (ggml-org#13062)

* cmake : do not include ./src as public for libllama

ggml-ci

* cmake : rework tests

ggml-ci

* llguidance : remove unicode include

ggml-ci

* cmake : make c++17 private

ggml-ci

arg : add --no-mmproj-offload (ggml-org#13093)

* arg : add --no-mmproj-offload

* Update common/arg.cpp