add encoding support for CSV files

[MonRani]

---

name: Pull Request
about: Create a pull request to contribute to the project
title: 'Fix CSV encoding error for ISO-8859-1/Latin-1 encoded files'
labels: 'bug-fix, csv, encoding'

Related Issue
Fixes #649

Description of Changes
This PR adds encoding support to CSV file loading to handle ISO-8859-1 (Latin-1) encoded files that were previously failing with "Invalid Input Error: CSV Error". The changes include:

• Added optional encoding parameter to CSVConfig with default "utf-8" for backward compatibility
• Updated CSVSource to pass encoding parameter to DuckDB's read_csv_auto function
• Modified configuration parsing to support encoding from TOML files
• Enhanced documentation with encoding parameter examples and usage instructions
• Added comprehensive test suite to verify encoding support

The fix maintains full backward compatibility while enabling support for legacy encoded files that contain special characters.

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• New example
• Test improvement

Testing

• Created test suite (simple_encoding_test.py) that verifies:
  • Default UTF-8 encoding works with regular files
  • UTF-8 encoding fails appropriately with Latin-1 files
  • Latin-1 encoding successfully loads Latin-1 files with special characters
• Verified backward compatibility with existing UTF-8 encoded files
• Tested with files containing special characters (José, François, Müller, etc.)

Usage Example
To load an ISO-8859-1 encoded CSV file, add to your preswald.toml:

[data.mobile_dataset]
type = "csv"
path = "data/mobile_dataset.csv"
encoding = "latin-1"