fix: 404 error on retrieving snippet content in symlinked dir

os.path.realpath also resolves symlinks. We only want to resolve
'..' to prevent escaping the snippet roots, os.path.abspath does
this.

This commit also adds some logging be able to better analyze other
potential problems.

Author: mariobuikhuizen

jupyterlab-snippets/loader.py

@@ -30,12 +30,15 @@ def get_snippet_content(self, snippet):
                 path = os.path.join(root_path, *snippet)
 
                 # Prevent access to the entire file system when the path contains '..'
-                accessible = os.path.realpath(path).startswith(root_path)
+                accessible = os.path.abspath(path).startswith(root_path)
+                if not accessible:
+                    print(f'jupyterlab-snippets: {path} not accessible from {root_path}')
 
                 if accessible and os.path.isfile(path):
                     with open(path) as f:
                         return f.read()
         except:
             raise tornado.web.HTTPError(status_code=500)
 
+        print(f'jupyterlab-snippets: {snippet} not found in {self.snippet_paths}')
         raise tornado.web.HTTPError(status_code=404)