Enhance spam filtering in /e/OS with improved learning, contact whitelisting, and user control options

Summary

Enhance the spam filtering system in /e/OS by improving the learning efficiency of Rspamd/other-spam-app, integrating contact whitelisting, and providing users with greater control over spam management.

Description

What is the feature?

This feature involves updating the current spam filter application (e.g., Rspamd) or replacing it with a more effective spam filtering app. The new or updated spam filter will include enhanced settings that allow users to:

• Fine-tune the sensitivity of the spam filter to better match their preferences.
• Enable automatic whitelisting of emails from contacts saved in their Nextcloud/murena.io address book.
• Access more granular spam filter settings, giving them greater control over how emails are classified and managed.

Who will use this new feature?

This feature is intended for all /e/OS users who manage their emails through the Murena webmail or the Mail app, particularly those who are frustrated with the current spam filtering inefficiencies.

Why these users would like to use this feature?

Users will appreciate this feature because it will:

• Improve the accuracy of the spam filter, reducing the number of legitimate emails wrongly flagged as spam.
• Simplify email management by ensuring important contacts are not mistakenly filtered.
• Give users more control over how spam is managed, enhancing their overall email experience.

Examples

• A user adds a contact to their Nextcloud address book, and future emails from this contact are automatically recognized as legitimate by Rspamd/other-spam-app.
• After marking a few emails as "not spam," Rspamd/other-spam-app adjusts more quickly to stop filtering similar emails as spam.
• A user adjusts the spam filter sensitivity to a lower setting, reducing the likelihood of important emails being flagged.

Reflection

Mockups

• N/A

Diagrams

• N/A

Validation

• Test that Rspamd/other-spam-app correctly whitelists contacts from the user's Nextcloud address book.
• Test that reducing the number of required actions leads to faster learning of spam preferences.
• Validate that users can effectively adjust the spam filter sensitivity and see the expected changes in email filtering behavior.