Llama.cpp updates checkpoint eviction to prevent duplicate entries.

The official source reports this update: b10864: server : apply checkpoint min-step eviction only when the checkpoint list is full. server : apply checkpoint min-step eviction only when the checkpoint list is full ( #28302 ) The spacing eviction in create_checkpoint() keeps the oldest checkpoint and erases every later one within checkpoint_min_step of it. For prompts shorter than checkpoint_min_step this drops the checkpoint at n_tokens - 4 that the next request resumes from, so hybrid/recurrent models re-prefill from the previous checkpoint instead.