fix: reuse token truncation helper

Maintainer edit: make token truncation include the suffix within the budget and route the consolidator through the shared helper so recent-history and archive truncation keep the same semantics.

tests/agent/test_consolidator.py

@@ -825,4 +825,4 @@ class TestArchiveTruncation:
         enc = tiktoken.get_encoding("cl100k_base")
         sent_content = mock_provider.chat_with_retry.call_args.kwargs["messages"][1]["content"]
         token_count = len(enc.encode(sent_content))
-        assert token_count <= 9_900 + 10  # small margin for truncation suffix
+        assert token_count <= 9_900

tests/agent/test_context_prompt_cache.py

@@ -237,8 +237,7 @@ def test_recent_history_truncated_at_max_tokens(tmp_path) -> None:
     assert len(history_section) == 2
 
     enc = tiktoken.get_encoding("cl100k_base")
-    # Small margin for the truncation suffix appended after the token slice.
-    assert len(enc.encode(history_section[1])) <= builder._MAX_HISTORY_TOKENS + 50
+    assert len(enc.encode(history_section[1])) <= builder._MAX_HISTORY_TOKENS
 
 
 def test_no_recent_history_when_dream_has_processed_all(tmp_path) -> None:

tests/utils/test_helpers.py

@@ -24,8 +24,7 @@ def test_truncate_text_to_tokens_truncates_over_budget():
     result = truncate_text_to_tokens(text, 50)
 
     assert result.endswith("\n... (truncated)")
-    body = result[: -len("\n... (truncated)")]
-    assert len(enc.encode(body)) <= 50
+    assert len(enc.encode(result)) <= 50
 
 
 def test_truncate_text_to_tokens_non_positive_budget_returns_text():