Support multiple GPUs (split mode) on SYCL backend (llama/5806)

Use batched mul_mat pathway (llama/5591)

Add support for soft_max ALiBi (llama/5639)

code : normalize enum names (llama/5697)

Introduce backend GUIDs (ggml/743)

conext add name (llama/5624)

Update ggml_sycl_op_mul_mat_vec_q (llama/5502)

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

src : relocate new backend sources