server : support unified cache across slots (#16736)

* server : support unified context across slots

* cont : fix speculative decoding initialization

* context : fix n_ctx_per_seq computation

* server : purge slots one by one

* tests : add unified cache server tests

* llama : update per-seq context computation

* test-thread-safety : handle tiny training context of the input model

* server : fix server_tokens clear()

* server : use 4 slots + unified KV by default

* llama : add note about context size queries

* cont : update todos [no ci]

* context : do not cap the size of the context

* tests : adjust parameters to be CI friendlier

* context : add warning

tools/server/tests/utils.py

@@ -78,6 +78,7 @@ class ServerProcess:
     server_embeddings: bool | None = False
     server_reranking: bool | None = False
     server_metrics: bool | None = False
+    kv_unified: bool | None = False
     server_slots: bool | None = False
     pooling: str | None = None
     draft: int | None = None
@@ -159,6 +160,8 @@ class ServerProcess:
             server_args.append("--reranking")
         if self.server_metrics:
             server_args.append("--metrics")
+        if self.kv_unified:
+            server_args.append("--kv-unified")
         if self.server_slots:
             server_args.append("--slots")
         else: