feat: unicode display width support

[Wybxc]

Compute unicode display width instead of string length.

Before:

🐱‍💻 Projects: 
计算机网络                    1 hr 58 mins        ████████░░░░░░░░░░░░░░░░░   32.72 % 
alphartc                 1 hr 51 mins        ████████░░░░░░░░░░░░░░░░░   30.67 % 
rust_learning_group_ans  1 hr 45 mins        ███████░░░░░░░░░░░░░░░░░░   29.21 % 
erg-prover               18 mins             █░░░░░░░░░░░░░░░░░░░░░░░░   05.12 % 
typst                    8 mins              █░░░░░░░░░░░░░░░░░░░░░░░░   02.28 % 

After:

🐱‍💻 Projects: 
计算机网络               1 hr 58 mins        ████████░░░░░░░░░░░░░░░░░   32.72 % 
alphartc                 1 hr 51 mins        ████████░░░░░░░░░░░░░░░░░   30.67 % 
rust_learning_group_ans  1 hr 45 mins        ███████░░░░░░░░░░░░░░░░░░   29.21 % 
erg-prover               18 mins             █░░░░░░░░░░░░░░░░░░░░░░░░   05.12 % 
typst                    8 mins              █░░░░░░░░░░░░░░░░░░░░░░░░   02.28 % 

Preview:

[Wybxc]

It appears that a previous CI failure is blocking this pull request. I will push a new commit to trigger the CI again and work on resolving the issue.

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

[Copilot]

The wcswidth function can return -1 when the string contains non-printable characters (e.g., control characters). This will cause incorrect behavior in the padding logic. Consider adding a check to handle this case, for example by falling back to using len(string) when wcswidth returns -1.

Suggested change

	width = wcswidth(string)
	width = wcswidth(string)
	if width < 0:
	# Fallback for strings with non-printable characters where wcswidth returns -1.
	width = len(string)