feat: checkpoint selection (#744)

* rebase selection code

* bug-free run: checkpoint selection and dynamic EDA loading

* add prototypes of various selectors, to imp. and test later

* fix EDA write bug

* move selector to from proposal.py tp seletc.py

* auto lint

* fix line-too-long typos

* aligh the design of "selection", rm extra instance check

* make auto-lint

* add non-trival selector: SOTAjump

rdagent/components/coder/data_science/workflow/__init__.py

@@ -59,7 +59,7 @@ class WorkflowMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
         # 2. code
         system_prompt = T(".prompts:workflow_coder.system").r(
             task_desc=workflow_information_str,
-            competition_info=self.scen.get_scenario_all_desc(),
+            competition_info=self.scen.get_scenario_all_desc(eda_output=workspace.file_dict.get("EDA.md", None)),
             queried_similar_successful_knowledge=queried_similar_successful_knowledge,
             queried_former_failed_knowledge=queried_former_failed_knowledge[0],
             out_spec=PythonAgentOut.get_spec(),

rdagent/components/coder/data_science/workflow/eval.py

@@ -127,7 +127,8 @@ class WorkflowGeneralCaseSpecEvaluator(CoSTEEREvaluator):
         stdout += "\n" + submission_check_out
 
         system_prompt = T(".prompts:workflow_eval.system").r(
-            scenario=self.scen.get_scenario_all_desc(),
+            # here we pass `None` to `eda_output` because we do not have nor need EDA output for workflow.
+            scenario=self.scen.get_scenario_all_desc(eda_output=None),
             task_desc=target_task.get_task_information(),
             spec=(
                 implementation.file_dict["spec/workflow.md"]