llama: add default load-mode auto, which avoids mmap on iGPUs (llama/26081)

llama: add default load-mode auto, which avoids mmap on iGPUs (llama/26081)

* llama: add new default load-mode auto which picks mmap unless a non-Metal iGPU is used

* Update ggml/src/ggml-hexagon/ggml-hexagon.cpp

Co-authored-by: Max Krasnyansky <redacted>
* set mmap_support to false on OpenCL backend

* fix order of load modes

* use -1 for auto

* resolve load mode auto earlier to correctly pick gpu host or cpu memory

* add load mode auto to llama-bench

* bump virtgpu api version, regenerate docs

---------

Co-authored-by: Piotr Wilkin (ilintar) <redacted>
Co-authored-by: Max Krasnyansky <redacted>
Co-authored-by: Georgi Gerganov <redacted>