fix: handle UnicodeDecodeError in PlainTextConverter when charset detection is inaccurate (fixes #1505)

[hanhan761]

Summary

When stream_info.charset is detected from partial file content (first 4096 bytes) and returns 'ascii', but the full file contains non-ASCII characters (e.g., Spanish accents, UTF-8 characters), .decode('ascii') raises UnicodeDecodeError.

This fix adds a fallback: if decoding with the detected charset fails, fall back to charset_normalizer for automatic charset detection on the full file content.

Issue

Fixes #1505

Verification

• UTF-8 files with Spanish characters now decode correctly
• Pure ASCII files continue to work as before
• Files where charset detection from partial content returns 'ascii' but full content has UTF-8 characters fall back gracefully