main : fix file existence check in main.cpp (#1889)

talk-llama : sync llama.cpp

make : fix CUBLAS link with WSL (#1878)

sync : ggml

ggml : resolve merge conflicts (ggml/0)

common : add IQ1_S (ggml/0)

ci : enable -Werror for CUDA builds (llama/5579)

cuda, metal : fix nans in soft_max (llama/5574)

ggml : android and old glibc NUMA incompatibility bugfixes (llama/5557)

ggml : restore vec dot stride arg names (llama/5453)

ci : fix wikitext url + compile warnings (llama/5569)

metal : fix unused warnings (llama/0)

ggml, common, examples, tests : fixed type arguments in printf (llama/5528)

1.5 bit quantization (llama/5453)

ggml : add ALiBi support for ggml_soft_max_ext (llama/5488)

ci : add an option to fail on compile warning (llama/3952)

cmake : fix VULKAN and ROCm builds (llama/5525)

ggml : add numa options (llama/5377)

cuda : print message when initialization fails (llama/5512)

vulkan: Find optimal memory type but with fallback (llama/5381)

Early return for zero size calls to get_tensor. (llama/5482)

ggml-quants : fix compiler warnings (shadow variable) (llama/5472)

ggml-sycl: Replace 3d ops with macro (llama/5458)

build : update CBLAS flags + fix unused var warning (#0)

main : check if input files exist before proceeding (#1872)

examples : clean up common code (#1871)

models : fix openvino setup info (#1874)

models : add update py requirements

swift : package no longer use ggml dependency (#1861)

whisper : fix external encoder (#1860)

sync : ggml

ggml-alloc : allocate all leafs as if they were inputs (ggml/731)

talk-llama : sync llama.cpp

sync : ggml

ggml-backend : sync remnant

CUDA: mul_mat_vec_q tiling, refactor mul mat logic (llama/5434)

vulkan: only use M-sized matmul on Apple GPUs (llama/5412)

ggml : fix compile warnings (unused vars) (llama/4966)

ggml : add mmla kernels for quantized GEMM (llama/4966)

metal : use autoreleasepool to avoid memory leaks (llama/5437)

ggml-alloc : v3 (ggml/727)

examples : added audio_ctx argument to main and server (#1857)

metal : option to embed MSL source into compiled binary (#1842)

examples : initialize context params properly (#1852)

talk-llama : sync llama.cpp

sync : ggml

src : relocate new backend sources

ggml : fix `error C2078: too many initializers` for MSVC ARM64 (llama/5404)

CUDA: more warps for mmvq on NVIDIA (llama/5394)

CUDA: fixed mmvq kernel for bs 2,3,4 and -sm row (llama/5386)