Drop storefront 1.0

[IKarbowiak]

Drop storefront 1.0 -> views, forms, templates, statics
Closes #5017
Closes #5046

Pull Request Checklist

1. Privileged views and APIs are guarded by proper permission checks.
2. All visible strings are translated with proper context.
3. All data-formatting is locale-aware (dates, numbers, and so on).
4. Database queries are optimized and the number of queries is constant.
5. Database migration files are up to date.
6. The changes are tested.
7. GraphQL schema and type definitions are up to date.
8. Changes are mentioned in the changelog.

[krzysztofwolski]

Dropping packages, dropping code. I love it.

[fowczarek]

I think we can remove:

• LOGIN_URL = "/account/login/"
• PAGINATE_BY = 16
• DASHBOARD_PAGINATE_BY = 30
• DASHBOARD_SEARCH_LIMIT = 5
• LOGOUT_ON_PASSWORD_CHANGE = False
• SOCIAL_AUTH_USERNAME_IS_FULL_EMAIL = True
• SOCIAL_AUTH_USER_MODEL = AUTH_USER_MODEL
• SOCIAL_AUTH_FACEBOOK_SCOPE = ["email"]
• SOCIAL_AUTH_FACEBOOK_PROFILE_EXTRA_PARAMS = {"fields": "id, email"}
• SOCIAL_AUTH_REDIRECT_IS_HTTPS = True
  form settings.py, and:
• "saleor.account.backends.google.CustomGoogleOAuth2",
• "saleor.account.backends.facebook.CustomFacebookOAuth2",
  from AUTHENTICATION_BACKENDS in settings.py.
  Can someone check settings too?
  @NyanKiyoshi @maarcingebala @krzysztofwolski @korycins

[NyanKiyoshi]

@IKarbowiak I pushed some changes to drop useless middleware, apps, context processors. They are not that heavy on the processing time but save between 15-25% of processing in my tests.

I also removed django-redis as it is no longer needed (no more session caching), django-social, and more–including npm dependencies that are no longer used.

[korycins]

Do we still need this? and docs?

[IKarbowiak]

We still use npm for build-emails, so probably node_modules must stay. But in case of docs for me, it looks like we can remove it. @maarcingebala, what do you think?

[maarcingebala]

Yes, this probably needs to stay for now since we still have some packages in package.json.

[korycins]

AFAIK this task was created to future usage from celery beat. But I am not sure. You can ask @derenio

[derenio]

This is true, although I'm not sure what's the status of shipping saleor with something like celery beat for which this task was designed. I think we can remove it. When we implement the default periodic tasks schema we will have to think about the minimal variant price recalculations.