WRITEUP

This patch addresses an integer overflow vulnerability in the gguf tensor size calculation in ggml.cpp. The fix adds a check to prevent overflow when accumulating the size of tensors, ensuring that the total size does not exceed SIZE_MAX.