Update GGML to b6646 (#12245)

Notable EOLs with this change:
- MacOS v12 and v13 are no longer supported (v14+ required)
- AMD gfx900 and gfx906 are no longer supported

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-f16.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q4_1.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_1);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q5_1.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_1);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-f16-q8_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q8_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-f16.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q4_1.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q5_1.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_0-q8_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-f16.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q4_1.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q5_1.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q4_1-q8_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-f16.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16);

ml/backend/ggml/ggml/src/ggml-cuda/template-instances/fattn-vec-f32-instance-hs128-q5_0-q4_0.cu

-// This file has been autogenerated by generate_cu_files.py, do not edit manually.
-
-#include "../fattn-vec-f32.cuh"
-
-DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0);