feat: integrate azure deepseek r1 (#591)

* fix several task & integrate deepseek R1

* fix CI

---------

Co-authored-by: Xu Yang <xuyang1@microsoft.com>

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

@@ -52,7 +52,7 @@ class FeatureCoSTEEREvaluator(CoSTEEREvaluator):
         de = DockerEnv(conf=ds_docker_conf)
 
         # TODO: do we need to clean the generated temporary content?
-        fname = "feature_test.py"
+        fname = "test/feature_test.py"
         test_code = (DIRNAME / "eval_tests" / "feature_test.txt").read_text()
         implementation.inject_files(**{fname: test_code})
 

rdagent/components/coder/data_science/feature/exp.py

@@ -7,4 +7,7 @@ from typing import Dict, Optional
 from rdagent.components.coder.CoSTEER.task import CoSTEERTask
 from rdagent.core.utils import cache_with_pickle
 
-FeatureTask = CoSTEERTask
+
+# Because we use isinstance to distinguish between different types of tasks, we need to use sub classes to represent different types of tasks
+class FeatureTask(CoSTEERTask):
+    pass