fix: Strip UTF-8 BOM in DataImportJob#csv_reader before parsing CSV (#14126)

## Description

`DataImportJob#csv_reader` reads CSV data with `force_encoding('UTF-8')`
but does not strip the UTF-8 Byte Order Mark (`EF BB BF`). If a CSV file
containing a BOM is imported, the first header key is prefixed with
`\uFEFF`, which causes key mismatches in `DataImport::ContactManager`
when the first column is one of the recognized keys (`:email`,
`:identifier`, `:phone_number`, `:name`).

This was identified during review of #14123 (see #14124 for the tracking
issue).

Fixes #14124

## Type of change

- [x] Bug fix (non-breaking change which fixes an issue)

## How Has This Been Tested?

- Added a new fixture (`spec/fixtures/data_import/with_bom.csv`)
containing a UTF-8 BOM followed by valid contact data.
- Added a new spec (`will strip UTF-8 BOM and import contacts
correctly`) that imports the BOM fixture and verifies that `name`,
`email`, and `phone_number` are all correctly parsed.
- All existing examples in `spec/jobs/data_import_job_spec.rb` continue
to pass.

## Checklist:

- [x] My code follows the style guidelines of this project
- [x] I have performed a self-review of my code
- [x] I have commented on my code, particularly in hard-to-understand
areas
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [x] I have added tests that prove my fix is effective or that my
feature works
- [x] New and existing unit tests pass locally with my changes
- [x] Any dependent changes have been merged and published in downstream
modules

app/jobs/data_import_job.rb

@@ -106,6 +106,7 @@ class DataImportJob < ApplicationJob
     raw_data = file.read
     utf8_data = raw_data.force_encoding('UTF-8')
     clean_data = utf8_data.valid_encoding? ? utf8_data : utf8_data.encode('UTF-16le', invalid: :replace, replace: '').encode('UTF-8')
+    clean_data = clean_data.delete_prefix("\xEF\xBB\xBF")
 
     CSV.new(StringIO.new(clean_data), headers: true)
   end

spec/fixtures/data_import/with_bom.csv

@@ -0,0 +1,2 @@
+name,email,phone_number
+Ahmed,ahmed@example.com,+971501234567

spec/jobs/data_import_job_spec.rb

@@ -91,6 +91,20 @@ RSpec.describe DataImportJob do
         expect(invalid_data_import.account.contacts.first.name).to eq(csv_data[0]['name'].encode('UTF-8', 'binary', invalid: :replace,
                                                                                                                     undef: :replace, replace: ''))
       end
+
+      it 'will strip UTF-8 BOM and import contacts correctly' do
+        bom_data_import = create(:data_import,
+                                 import_file: Rack::Test::UploadedFile.new(Rails.root.join('spec/fixtures/data_import/with_bom.csv'),
+                                                                           'text/csv'))
+
+        described_class.perform_now(bom_data_import)
+        expect(bom_data_import.account.contacts.count).to eq(1)
+
+        contact = bom_data_import.account.contacts.first
+        expect(contact.name).to eq('Ahmed')
+        expect(contact.email).to eq('ahmed@example.com')
+        expect(contact.phone_number).to eq('+971501234567')
+      end
     end
 
     context 'when the data contains existing records' do