server: update readme to mention n_past_max metric (#16436)

https://github.com/ggml-org/llama.cpp/pull/15361 added new metric
exported, but I've missed this doc.

diff --git a/tools/server/README.md b/tools/server/README.md
index 9f7ab229f7ddf5594e61a599bb28b29125116bc6..6825c8bf300c6887504b251cffc03587a0a35966 100644 (file)
--- a/tools/server/README.md
+++ b/tools/server/README.md
@@ -1045,6 +1045,7 @@ Available metrics:
 - `llamacpp:kv_cache_tokens`: KV-cache tokens.
 - `llamacpp:requests_processing`: Number of requests processing.
 - `llamacpp:requests_deferred`: Number of requests deferred.
+- `llamacpp:n_past_max`: High watermark of the context size observed.
 
 ### POST `/slots/{id_slot}?action=save`: Save the prompt cache of the specified slot to a file.