Accept string arguments in SequenceSet delete?

[OskarEichler]

Summary

Use String#match? when deciding whether a delete? argument represents one number or a range. String#include? does not accept a Regexp, so documented string inputs currently raise TypeError. Retain the numeric-versus-SequenceSet return distinction introduced by #583.

Reproduction

require 'net/imap'
set = Net::IMAP::SequenceSet.new('1:5')
p set.delete?('2') # 2
p set.delete?('3:4').to_s # "3:4"
p set.to_s # "1,5"
# Before: TypeError on the first deletion.

Verification

• 39 focused checks; 28 failing expectations before, zero afterward. Empty/nonempty sets, present/absent numbers, reversed/equal ranges, star and frozen rejection are covered. The source of merged 🐛 Fix SequenceSet#delete?(num..num) to return set #583 and its earlier numeric/range return policy was reviewed.
• Existing rake test on this isolated branch: 1726 tests, 12604 assertions, 0 failures, 0 errors, 0 pendings, 0 omissions, 0 notifications, Ruby 4.0.6 via rbenv. The baseline also passes all 1,726 tests; assertion counts vary slightly across runs.
• Supplemental RuboCop Lint retains the same 48 existing findings. Ruby syntax and git diff --check pass. No repository tests, dependencies or workflows were added or modified; focused reproductions live outside the repository under the consumer's no-new-tests policy.
• Independent branch based on 6d2ef7a636a1e2449187a83b06ac7a5baa54ead2; the runtime diff from released 0.6.6 is documentation-only before this patch.

Compatibility and limits

No public signature change. String numbers now return a number and string ranges return a SequenceSet, as intended. Missing entries still return nil and frozen sets still raise FrozenError. Other Ruby/OS versions were not run locally. No production or external IMAP service was used. Local verification does not imply upstream CI approval or comprehensive behavioral coverage.

[nevans]

please add a test