tasks: normalize the url argument name of most lister

Since all the listing tasks accepts an url as first argument (whatever the
argument name is), it makes sense to use a simple common argument name for
this. I've chosen 'url' instead of api_baseurl/forge_url/url.

Also kill now useless `new_lister()` functions.

swh/lister/npm/lister.py

@@ -16,10 +16,9 @@ class NpmListerBase(IndexingHttpLister):
     LISTER_NAME = 'npm'
     instance = 'npm'
 
-    def __init__(self, api_baseurl='https://replicate.npmjs.com',
+    def __init__(self, url='https://replicate.npmjs.com',
                  per_page=1000, override_config=None):
-        super().__init__(api_baseurl=api_baseurl,
-                         override_config=override_config)
+        super().__init__(url=url, override_config=override_config)
         self.per_page = per_page + 1
         self.PATH_TEMPLATE += '&limit=%s' % self.per_page
 
@@ -76,7 +75,7 @@ class NpmListerBase(IndexingHttpLister):
             'https://www.npmjs.com/package/%s' % repo_name,
             # package metadata url needs to be escaped otherwise some requests
             # may fail (for instance when a package name contains '/')
-            '%s/%s' % (self.api_baseurl, quote(repo_name, safe=''))
+            '%s/%s' % (self.url, quote(repo_name, safe=''))
         )
 
     def string_pattern_check(self, inner, lower, upper=None):

swh/lister/npm/tasks.py

@@ -14,7 +14,7 @@ from swh.lister.npm.models import NpmVisitModel
 @contextmanager
 def save_registry_state(lister):
     params = {'headers': lister.request_headers()}
-    registry_state = lister.session.get(lister.api_baseurl, **params)
+    registry_state = lister.session.get(lister.url, **params)
     registry_state = registry_state.json()
     keys = ('doc_count', 'doc_del_count', 'update_seq', 'purge_seq',
             'disk_size', 'data_size', 'committed_update_seq',