feat: add job to remove stale contacts and contact_inboxes (#11186)

- Add a job to remove stale contacts and contact_inboxes across all accounts

Stale anonymous contact is defined as 
- have no identification (email, phone_number, and identifier are NULL)
- have no conversations
- are older than 30 days

---------

Co-authored-by: Pranav <pranav@chatwoot.com>
Co-authored-by: Shivam Mishra <scm.mymail@gmail.com>

app/jobs/internal/process_stale_contacts_job.rb

@@ -0,0 +1,18 @@
+# housekeeping
+# remove stale contacts for all accounts
+# - have no identification (email, phone_number, and identifier are NULL)
+# - have no conversations
+# - are older than 30 days
+
+class Internal::ProcessStaleContactsJob < ApplicationJob
+  queue_as :scheduled_jobs
+
+  def perform
+    Account.find_in_batches(batch_size: 100) do |accounts|
+      accounts.each do |account|
+        Rails.logger.info "Enqueuing RemoveStaleContactsJob for account #{account.id}"
+        Internal::RemoveStaleContactsJob.perform_later(account)
+      end
+    end
+  end
+end

app/jobs/internal/remove_stale_contacts_job.rb

@@ -0,0 +1,13 @@
+# housekeeping
+# remove contacts that:
+# - have no identification (email, phone_number, and identifier are NULL)
+# - have no conversations
+# - are older than 30 days
+
+class Internal::RemoveStaleContactsJob < ApplicationJob
+  queue_as :low
+
+  def perform(account, batch_size = 1000)
+    Internal::RemoveStaleContactsService.new(account: account).perform(batch_size)
+  end
+end

app/models/contact.rb

@@ -128,6 +128,18 @@ class Contact < ApplicationRecord
     )
   }
 
+  # Find contacts that:
+  # 1. Have no identification (email, phone_number, and identifier are NULL or empty string)
+  # 2. Have no conversations
+  # 3. Are older than the specified time period
+  scope :stale_without_conversations, lambda { |time_period|
+    where('contacts.email IS NULL OR contacts.email = ?', '')
+      .where('contacts.phone_number IS NULL OR contacts.phone_number = ?', '')
+      .where('contacts.identifier IS NULL OR contacts.identifier = ?', '')
+      .where('contacts.created_at < ?', time_period)
+      .where.missing(:conversations)
+  }
+
   def get_source_id(inbox_id)
     contact_inboxes.find_by!(inbox_id: inbox_id).source_id
   end

app/services/internal/remove_stale_contacts_service.rb

@@ -0,0 +1,19 @@
+class Internal::RemoveStaleContactsService
+  pattr_initialize [:account!]
+
+  def perform(batch_size = 1000)
+    contacts_to_remove = @account.contacts.stale_without_conversations(30.days.ago)
+    total_deleted = 0
+
+    Rails.logger.info "[Internal::RemoveStaleContactsService] Starting removal of stale contacts for account #{@account.id}"
+
+    contacts_to_remove.find_in_batches(batch_size: batch_size) do |batch|
+      contact_ids = batch.map(&:id)
+
+      ContactInbox.where(contact_id: contact_ids).delete_all
+      Contact.where(id: contact_ids).delete_all
+      total_deleted += batch.size
+      Rails.logger.info "[Internal::RemoveStaleContactsService] Deleted #{batch.size} contacts (#{total_deleted} total) for account #{@account.id}"
+    end
+  end
+end