metal : support FA without mask + add asserts (llama/7278)

ggml : resolve merge (ggml/0)

ggml : full ALiBi support (llama/7192)

metal : fix unused warning

ggml : add Flash Attention (llama/5021)

ggml : group all experts in a single ggml_mul_mat_id (llama/6505)

llama : add qwen2moe (llama/6074)

Added support for GGML_OP_CLAMP in Metal (llama/6662)

metal : unify mul_mv_id kernels (llama/6556)

feat: implemented sigmoid function (ggml/806)

ggml : mul_mat_id use the same tensor for all the experts (llama/6387)

sync : ggml (#2001)

metal : build metallib + fix embed path (llama/6015)

ggml : reuse quantum structs across backends (llama/5943)

1.5 bit: we can do even better (llama/5999)

Better 1.5 bit quantization (llama/5971)

metal : move mm_id indices to shared mem (llama/5982)

ggml : add ggml-common.h to deduplicate shared code (llama/5940)

ggml : IQ3_S improvements (llama/5829)

add some new ops, fix some operators and add batch operations to certain operators. (ggml/747)

ggml : make i-quants work with super-blocks of 64 (CPU,Metal) (llama/5760)

IQ4_XS: a 4.25 bpw quantization (llama/5747)

Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (llama/5721)

IQ3_S: a much better alternative to Q3_K (llama/5676)

sync : llama.cpp (ggml/0)

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

metal : fix unused warnings (llama/0)

1.5 bit quantization (llama/5453)

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

metal : add im2col F32 dst support (llama/5132)

ggml : fix IQ3_XXS on Metal (llama/5219)

SOTA 3-bit quants (llama/5196)

ggml : SOTA 2-bit quants (add IQ2_XS) (llama/4856)

metal : improve dequantize precision to match CPU (llama/4836)

SOTA 2-bit quants (llama/4773)

metal : add kernel_get_rows_i32

metal : optimize ggml_mul_mat_id (faster Mixtral PP) (llama/4725)

metal : enable shader debugging (cmake option) (llama/4705)

sync : ggml (ggml_scale, ggml_row_size, etc.) (#1677)

sync : ggml (Metal fixes, new ops, tests) (#1633)

sync : ggml (new ops, new backend, etc) (#1602)

whisper : add full CUDA and Metal offloading (#1472)

sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)

metal : add F32 support + update bench output

whisper : Metal and ggml-alloc support (#1270)

sync : ggml (HBM + Metal + style) (#1264)

ggml : sync latest llama.cpp (view_src + alloc improvements) (#1247)

ggml : sync (ggml-alloc, GPU, eps, etc.) (#1220)

ggml : sync latest repo (mostly refactoring changes)

metal : sync ggml-metal (ref #1047)