Fix Segment._split_cells for non-unit-width characters (emoji, CJK)

[ShiLiguo123]

Summary

Fix Segment._split_cells() to correctly handle strings containing double-cell-width characters (emoji, CJK characters, etc.).

Root Cause

The old algorithm used a heuristic initial position: pos = int((cut / cell_length) * len(text)). This assumes uniform character width, which fails badly for mixed-width strings.

For example:

s = Segment('🦊🦊🦊\n\n\n\n\n\n')
Segment._split_cells(s, 3)
# Wrong: (Segment('🦊🦊🦊\n '), Segment(' \n\n\n\n'))

Fix

Replaced the heuristic + while-loop with a character-by-character walk-through. The new algorithm:

1. Tracks cumulative cell position as it iterates over characters
2. When it finds the character that exceeds the cut point, it splits there
3. Double-width characters are replaced with two spaces as before

Fixes #3299

[willmcgugan]

Your PR has been closed due to a AI policy violation.

Please read the following before submitting further PRs.

https://github.com/Textualize/textual/blob/main/AI_POLICY.md