feat: enhance timeout management and knowledge base handling in CoSTEER components (#1130)

* feat: enhance timeout management and knowledge base handling in CoSTEER components

* fix a little bug

* fix small bug

* fix a small bug

* Update rdagent/scenarios/data_science/loop.py

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

* add scale check

* fix a small bug

* fix CI

* use dynamic chat_token_limit & remove repeated lines

* fix CI

* remove useless comment

* fix small bug

* update draft appendix

* fix prompt

* add code correctness as top priority

---------

Co-authored-by: you-n-g <you-n-g@users.noreply.github.com>

rdagent/components/coder/model_coder/eva_utils.py

@@ -87,7 +87,7 @@ class ModelCodeEvaluator(CoSTEEREvaluator):
                     user_prompt=user_prompt,
                     system_prompt=system_prompt,
                 )
-                > LLM_SETTINGS.chat_token_limit
+                > APIBackend().chat_token_limit
             ):
                 execution_feedback_to_render = execution_feedback_to_render[len(execution_feedback_to_render) // 2 :]
             else:
@@ -142,7 +142,7 @@ class ModelFinalEvaluator(CoSTEEREvaluator):
                     user_prompt=user_prompt,
                     system_prompt=system_prompt,
                 )
-                > LLM_SETTINGS.chat_token_limit
+                > APIBackend().chat_token_limit
             ):
                 execution_feedback_to_render = execution_feedback_to_render[len(execution_feedback_to_render) // 2 :]
             else:

rdagent/components/coder/model_coder/evolving_strategy.py

@@ -61,7 +61,7 @@ class ModelMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
                     user_prompt=user_prompt,
                     system_prompt=system_prompt,
                 )
-                < LLM_SETTINGS.chat_token_limit
+                < APIBackend().chat_token_limit
             ):
                 break
             elif len(queried_former_failed_knowledge_to_render) > 1: