Fix: Correct Content-Length for StringIO with multi-byte characters

[hcthakur2004]

Summary

Fixes a regression where Content-Length is incorrectly calculated for io.StringIO objects containing multi-byte characters (like emojis).

The Issue

When io.StringIO is passed as data, super_len uses character count via len or tell() instead of the UTF-8 encoded byte length. This mismatch causes server errors when the encoded body is larger than the reported length.

The Fix

Updated super_len in utils.py to encode StringIO content to UTF-8 before calculating length, matching the behavior for standard strings and urllib3 expectations.