feat: support mmap for model loading

[wbruna]

Introduces a new --use-mmap flag that replaces model loading I/O operations with mmap + memcpy.

In my tests, this helps model loading speed slightly, though the gain was never higher than half a second. Its primary benefit right now is validation of the mmap backend implementation. Later, I plan to extend this to allow the mapped file to serve directly as weight storage for backends that use main memory.

I used a non-default flag to be extra safe, but we could arguably follow llama.cpp approach, with a --no-mmap flag to disable it instead.

I was only able to test (and build...) it under Linux, so additional testing is very welcome 🙂