Tracking (storage): support state cleaning on non-pk prefix watermark column

[hzxa21]

We decided to adopt option 3 mentioned here

[github-actions]

This issue has been open for 60 days with no activity.

If you think it is still relevant today, and needs to be done in the near future, you can comment to update the status, or just manually remove the no-issue-activity label.

You can also confidently close this issue as not planned to keep our backlog clean.
Don't worry if you think the issue is still valuable to continue in the future.
It's searchable and can be reopened when it's time. 😄

[Li0k]

Disscuss wtih @hzxa21 and @st1page, i found some assumptions exist for non_pk_watermark

1. the length of the watermark col is 1.
2. the state table that needs to be cleaned, the watermrak_col exists in pk.
3. the operator guarantees not to access data below the watermark, storage does not need to give this guarantee.

Therefore, the design on the storage side is as follows:

1. state table holds watermark_col
2. serialise watermark based on watermark_col instead of prefix(1)
3. when writing to TableWaterMark, record that it is a non_pk_watermark
4. modify the skip_watermark_iter implementation so that non_pk_watermark needs to solve for watermark_col before comparing.