fix(chat): stabilize system prompt, sequence memory extraction, and send stable session id to preserve KV cache (#3360)

* fix(chat): stabilize system prompt, sequence memory extraction, send stable session id to preserve KV cache

Fixes #2927. As diagnosed in the issue, three things in Odysseus's request
pattern actively destroyed local backends' (llama.cpp / LM Studio) KV-cache
continuity, forcing a full prompt re-evaluation (15-30s+) on every turn:

1. Dynamic content folded into the system prompt every turn. Both the chat
   preface (ChatProcessor.build_context_preface) and the agent system prompt
   (_build_system_prompt) injected current_datetime_prompt() — text that
   changes every minute — directly into system-role messages, which llm_core
   then concatenates into the single system message sent as the cached
   prefix. Any byte difference there invalidates the entire cache. Moved this
   to a new current_datetime_context_message() helper that returns a
   standalone user-role message, inserted near the end of the array (right
   before the latest user turn) instead of mixed into the system prompt. The
   static system prefix (preset prompt + safety policy + agent base prompt)
   now stays byte-identical across turns of the same session.

2. Memory/skill extraction side-requests competed with the main completion.
   run_post_response_tasks fired extract_and_store / maybe_extract_skill via
   asyncio.create_task — fire-and-forget coroutines that could overlap the
   next turn's main request and steal llama.cpp's limited processing slots,
   evicting the cached checkpoint. They're now queued through a new
   _run_extraction_jobs_sequentially helper that waits for the session's
   stream to go idle and runs the jobs strictly one at a time.

3. No stable session identifier was sent to local backends, so llama.cpp
   assigned a new processing slot via LRU every turn ("session_id=<empty>
   server-selected (LCP/LRU)"), losing slot affinity. Added
   _apply_local_cache_affinity() in llm_core, which sets session_id and
   cache_prompt: true on outgoing payloads — gated to self-hosted
   OpenAI-compatible endpoints only (never api.openai.com or other cloud
   providers, which reject unrecognized request fields with a 400). Threaded
   session_id through stream_llm / llm_call_async / stream_agent_loop from
   the existing Odysseus session id.

Tests in tests/test_kv_cache_invalidation_2927.py exercise the real payload-
assembly and scheduling code paths: byte-identical system prefix across two
turns of the same session (with a regression check that genuinely changed
instructions DO still change it), the dynamic time block landing as a
user-role message, extraction jobs waiting for the stream to go idle and
running sequentially, and the outgoing payload carrying a stable session_id
(same across turns of one session, different across sessions) only for
self-hosted endpoints. Updated tests/test_user_time.py for the new message
placement.

* fix(tests): accept owner= kwarg in normalize_model_id monkeypatch

The upstream normalize_model_id signature now takes an owner= keyword
argument, and chat_helpers.py passes owner=getattr(sess, "owner", None)
at the call site. Update the test stub lambda to **kwargs so it handles
the new argument without breaking, and update chat_helpers.py to forward
the owner parameter consistently.

---------

Co-authored-by: Alexandre Teixeira <111787685+alteixeira20@users.noreply.github.com>

src/chat_processor.py

@@ -175,6 +175,19 @@ class ChatProcessor:
 
         Returns:
             Tuple of (preface messages, rag_sources list)
+
+        Note on KV-cache friendliness: the ``system``-role messages assembled
+        here are later concatenated into a single system message and sent as
+        the very first thing in the payload (see ``llm_core``'s "consolidate
+        system messages" step). Local OpenAI-compatible backends (llama.cpp /
+        LM Studio) key their KV cache off the byte-identical token prefix, so
+        *anything* that changes turn-to-turn — timestamps, retrieved snippets,
+        per-turn counts — must NOT be folded into a system message here. Such
+        content belongs in a separate ``user``/context message appended near
+        the end of the array (see ``current_datetime_context_message`` and
+        ``untrusted_context_message`` callers in ``build_chat_context``),
+        which keeps the static system prefix byte-identical across turns of
+        the same session and lets the backend reuse its cached prefix.
         """
         preface = []
         rag_sources = []
@@ -185,15 +198,6 @@ class ChatProcessor:
                 "role": "system",
                 "content": preset_system_prompt
             })
-        if not agent_mode:
-            try:
-                from src.user_time import current_datetime_prompt
-                preface.append({
-                    "role": "system",
-                    "content": current_datetime_prompt(),
-                })
-            except Exception:
-                logger.debug("Failed to add current date/time context", exc_info=True)
         preface.append({
             "role": "system",
             "content": UNTRUSTED_CONTEXT_POLICY,