Add Mistral, Qwen, Gemini support; Fix minor bugs

desktop_env/envs/desktop_env.py

@@ -30,7 +30,7 @@ def _execute_command(command: List[str]) -> None:
         p = subprocess.Popen(command)
         p.wait()
     else:
-        result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, timeout=60, text=True)
+        result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, timeout=60, text=True, encoding="utf-8")
         if result.returncode != 0:
             raise Exception("\033[91m" + result.stdout + result.stderr + "\033[0m")
         return result.stdout

desktop_env/evaluators/metrics/gimp.py

@@ -328,6 +328,9 @@ def check_structure_sim(src_path, tgt_path):
     Check if the structure of the two images are similar
     gimp:2a729ded-3296-423d-aec4-7dd55ed5fbb3
     """
+    if src_path is None or tgt_path is None:
+        return 0.
+
     img_src = Image.open(src_path)
     img_tgt = Image.open(tgt_path)
     structure_same = structure_check_by_ssim(img_src, img_tgt)