fix: handle mixed dictionary/plain ArrowScan batches

[GayathriSrividya]

closes #3260

Summary

• decode dictionary-encoded columns in ArrowScan.to_table before concatenation
• keep permissive concat behavior for other schema variations
• add a regression test covering mixed plain-string and dictionary-encoded batches

Testing

• /opt/homebrew/bin/ruff format pyiceberg/io/pyarrow.py tests/io/test_pyarrow.py
• /opt/homebrew/bin/ruff check --fix pyiceberg/io/pyarrow.py tests/io/test_pyarrow.py
• PYTHONPATH=. uv run --project . pytest tests/io/test_pyarrow.py -k mixed_dictionary_and_plain_string_batches -xvs

[github-actions]

This pull request has been marked as stale due to 30 days of inactivity. It will be closed in 1 week if no further activity occurs. If you think that's incorrect or this pull request requires a review, please simply write any comment. If closed, you can revive the PR at any time and @mention a reviewer or discuss it on the dev@iceberg.apache.org list. Thank you for your contributions.