common : Add a warning when we can't match samplers from a string or …

cuda : remove nrows_x in mul_mat_q_process_tile (#13325)

Signed-off-by: Xiaodong Ye <[email protected]>

examples : remove infill (#13283)

ggml-ci

llama : support tie embedding for chatglm models (#13328)

CUDA: mix virt/real CUDA archs for GGML_NATIVE=OFF (#13135)

clip : refactor graph builder (#13321)

* mtmd : refactor graph builder

* fix qwen2vl

* clean up siglip cgraph

* pixtral migrated

* move minicpmv to a dedicated build function

* move max_feature_layer to build_llava

* use build_attn for minicpm resampler

* fix windows build

* add comment for batch_size

* also support tinygemma3 test model

* qwen2vl does not use RMS norm

* fix qwen2vl norm (2)

sampling : make top_n_sigma no-op at <=0 or a single candidate (#13345)

sampling : don't consider -infinity values in top_n_sigma (#13344)

SYCL: Disable reorder optimize by default and stop setting tensor ext…

llama : fix build_ffn without gate (#13336)

* llama : fix build_ffn without gate

* fix build on windows

* Revert "fix build on windows"

This reverts commit fc420d3c7eef3481d3d2f313fef2757cb33a7c56.