Fix failing --no-check-version cli option

Currently, if the --no-check-version option is used,
the scan fails because an unbound variable is passed to
the run_scan function.
Fix this by assigning None in that case. Add a test that
passes with this change and fails without it.

Signed-off-by: alexzurbonsen <alexander.zur.bonsen@tngtech.com>

Author: alexzurbonsen

src/scancode/cli.py

@@ -480,6 +480,8 @@ def scancode(
         if check_version:
             from scancode.outdated import check_scancode_version
             outdated = check_scancode_version()
+        else:
+            outdated = None
 
         # run proper
         success, _results = run_scan(

tests/scancode/test_cli.py

@@ -121,6 +121,12 @@ def test_run_scan_includes_outdated_in_extra():
     assert results['headers'][0]['extra_data']['OUTDATED'] == 'out of date'
 
 
+def test_no_version_check_run_is_successful():
+    test_file = test_env.get_test_loc('single/iproute.c')
+    result_file = test_env.get_temp_file('json')
+    run_scan_click(['--no-check-version', test_file, '--json', result_file], expected_rc=0)
+
+
 def test_usage_and_help_return_a_correct_script_name_on_all_platforms():
     result = run_scan_click(['--help'])
     assert 'Usage: scancode [OPTIONS]' in result.output