Improve compatibilities with some column types

[lukasmasuch]

Describe your changes

This PR improves the compatibility related to some column types when serialized to parquet (instead of arrow). This includes:

• Converts columns containing interval, period, set, tuple and datetime object types to string.
• Uses the newer string type for better null support.
• Converts all non-string column names to string
• Uses column config to configure certain types correctly for editing: categorical, text, and boolean.
• Auto converts date, time, and datetime object columns to pandas native datetime columns to support editing.
• Support byte arrays as list input. Parquet interprets number lists as byte arrays.
• Always lowercase the type name, parquet sometimes uses uppercase type names (e.g. Uint64) which are not detected correctly.

---

Contribution License Agreement

By submitting this pull request you agree that all contributions to this project are made under the Apache 2.0 license.

[whitphx]

Thank you very very much! I really appreciate your support and this PR filling the edge cases I couldn't cover 😸