Use strict=False when reading repo files

In some cases they may contain duplicate sections. With it set to False
the last one wins instead of causing a traceback.

Related: rhbz#1678937

src/pylorax/api/projects.py

@@ -553,7 +553,7 @@ def get_source_ids(source_path):
     if not os.path.exists(source_path):
         return []
 
-    cfg = ConfigParser()
+    cfg = ConfigParser(strict=False)
     cfg.read(source_path)
     return cfg.sections()
 
@@ -587,7 +587,7 @@ def delete_repo_source(source_glob, source_name):
     found = False
     for f in glob(source_glob):
         try:
-            cfg = ConfigParser()
+            cfg = ConfigParser(strict=False)
             cfg.read(f)
             if source_name in cfg.sections():
                 found = True

tests/pylorax/repos/duplicate-test.repo

@@ -0,0 +1,23 @@
+[duplicate-repo]
+name=Duplicate repo #1
+failovermethod=priority
+baseurl=file:///tmp/lorax-empty-repo/
+enabled=1
+metadata_expire=7d
+repo_gpgcheck=0
+type=rpm
+gpgcheck=1
+gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-$releasever-$basearch
+skip_if_unavailable=False
+
+[duplicate-repo]
+name=Duplicate repo #2
+failovermethod=priority
+baseurl=file:///tmp/lorax-empty-repo/
+enabled=1
+metadata_expire=7d
+repo_gpgcheck=0
+type=rpm
+gpgcheck=1
+gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-$releasever-$basearch
+skip_if_unavailable=False