Match ID string literal bytes without encoding conflicts

[OskarEichler]

Summary

Match the literal-selection regexp against a binary view of ID strings. Matching its high-byte range directly against UTF-8 raises Encoding::CompatibilityError before ClientID can send ordinary non-ASCII names or values. Preserve the original bytes and the existing quoted-versus-literal policy.

Reproduction

require 'net/imap'
value = Net::IMAP::StringFormatter.string('日本語')
p [value.class, value.data.bytesize] # [Net::IMAP::Literal, 9]
# Before: Encoding::CompatibilityError.

Verification

• 92 focused checks; 22 failing expectations before, zero afterward. UTF-8, ISO-8859-1, binary and invalid UTF-8 byte strings, frozen/mutable inputs, ASCII quoting, CR/LF, nil and symbols, plus ClientID key/value formatting are covered.
• Existing rake test on this isolated branch: 1726 tests, 12594 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 signature or wire-format change for previously working input. Non-ASCII strings are sent as literals without transcoding. This is a formatting fix, not a change to charset negotiation or ID input validation. 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]

I have some other work in progress that will conflict with (clobber) this. But I'll either merge yours first (so it can go out in the next release) or merge mine into this PR (and give you the credit 😉).

[nevans]

Even though I'll probably clobber your implementation with mine, I might not finish mine before the next release. Your bugfix is tiny and focused, and fixes the bug. Mine is part of a bigger refactoring of how Net::IMAP handles arguments and compiles commands.

So, can you add a small test for this, so it can go out with the next release? Thanks!

[OskarEichler]

Added a focused regression test covering non-ASCII text through StringFormatter.string. The full command-data test file passes: 59 tests, 427 assertions.

[nevans]

Thanks!