Don't use capture_output=True with subprocess.run()

The `capture_output` argument does not exist on Python 3.6 used by
default on RHEL-8.

Signed-off-by: Tomáš Hozza <thozza@redhat.com>

test/run/test_mount.py

@@ -163,7 +163,7 @@ def test_all_options(tmpdir):
                 assert st.st_gid == 0
 
                 shortname_tested = False
-                proc = subprocess.run("mount", capture_output=True, check=True)
+                proc = subprocess.run("mount", stdout=subprocess.PIPE, check=True)
                 for line in proc.stdout.splitlines():
                     strline = line.decode("utf-8")
                     if mountpoint in strline: