[codex] Avoid ranged octet string index overflow

[jadamcrain]

Summary

Avoid a debug-build integer overflow when iterating ranged octet string data whose final index is u16::MAX.

The iterator already captured the correct index before incrementing, so release builds wrapped after the last item and still produced the expected output. This change makes the increment explicit and consistent with the generic ranged iterator by using a saturating increment.

It is not a production (release) issue.

[gemini-code-assist]

Code Review

This pull request enhances the safety of the DNP3 parser by replacing a standard increment with saturating_add in the RangedBytesIterator, preventing potential overflow panics when the index reaches its maximum value. A corresponding test case was added to verify parsing at the u16::MAX boundary. The reviewer noted that for full consistency and safety, the Index trait implementation for u16 should also be updated to use saturating arithmetic, as it currently still uses a standard increment that could trigger panics in debug builds.