ggml-org/llama.cpp b7819

on GitHub

graph : utilize ggml_build_forward_select() to avoid reallocations (#18898)

• graph : avoid branches between embedding and token inputs

• models : make deepstack graphs (e.g. Qwen3 VL) have constant topology

• ci : enable -DGGML_SCHED_NO_REALLOC=ON for server CI

• cont : pad token embeddings to n_embd_inp