ENH: Add engine='polars' support in read_csv

[abujabarmubarak]

🚀 Enhancement: Add engine='polars' Support in read_csv

🔧 Summary of Changes

This PR introduces support for using [Polars](https://pola-rs.github.io/polars/py-polars/html/reference/api/pl.read_csv.html) as a backend CSV parsing engine in pandas.read_csv, providing faster parsing capabilities for large files.

The following changes are included:

• ✅ Added support for engine="polars" in pandas.read_csv

• ✅ Dynamically imported Polars and handled ImportError gracefully

• ✅ Filtered read_csv() kwargs to only allow those compatible with Polars

• ✅ Converted Path input to string (Polars does not accept path-like objects in all versions)

• ✅ Added test case test_read_csv_with_polars under tests/io/parser

• ✅ Updated version to 2.3.3.dev0 in __init__.py and pyproject.toml (as part of the development build)

• ✅ Resolved all ruff linter errors and pre-commit hook failures (e.g., B904, E501, F841, SC1017)

• ✅ Formatted shell scripts using dos2unix to fix line-ending issues across:

  • ci/code_checks.sh
  • ci/run_tests.sh
  • scripts/cibw_before_build.sh
  • scripts/download_wheels.sh
  • scripts/upload_wheels.sh
  • gitpod/workspace_config

---

📆 Usage Example

import pandas as pd

df = pd.read_csv("sample.csv", engine="polars")
print(df)

✅ Expected Output:

   a  b
0  1  2
1  3  4

---

💡 Why This Matters

Polars is a high-performance DataFrame library designed for speed and multi-threaded performance. Adding it as a supported backend:

• Provides significant performance boosts for CSV reading
• Enhances flexibility for end-users to choose engines (like c, python, or polars)
• Keeps Pandas future-ready with optional modular parsing backends

---

✅ Tests & Quality Checks

• 🔪 Unit test added: test_read_csv_with_polars
• ✅ Passed: All pytest tests
• ✅ Passed: All pre-commit hooks
• ✅ Passed: ruff, shellcheck, cython-lint, codespell, etc.
• ↺ Converted scripts to LF line endings using dos2unix for consistent CI/CD compatibility

---

🧠 Notes

• polars is treated as an optional dependency
• If not installed, Pandas will raise a clear error:
  “Polars is not installed. Please install it with 'pip install polars'.”

---

🙌 Acknowledgements

Thanks to the maintainers for reviewing this contribution!
Looking forward to feedback or further improvements.