One day back offset for each date

### Bug description

The bug is frontend-implemented offset for date from database:
When selecting via SQL Lab, dates are from 1st to 31st august.
When creating table chart - dates are from 31st July to 30th August.
This is happening only while rendering chart, due to broken timezone qualification in Windows.
How to reproduce:
1. Check if your windows timezone is TIME_ZONE_ID_UNKNOWN via powershell command "w32tm /tz"
2. Create any dataset (greenplum/postgresql verified by now) with date
3. Create table chart with that date - see that dates have changed
4. go to F12 - three dots - more tools - sensors - TZ override - see that dates became correct

Workaround is to setup "hours offset" in dataset properties to 3 hours.
Another workaround is to remove checkbox "is temporal" from date column in dataset.
But we are facing migration of hundreds datasets and these are not automatized productive solution.

I have found almost relevant issue fixed in https://github.com/apache/superset/pull/42329 but it is backend props and I have no qualification to confirm it would help my frontend.

### Screenshots/recordings

### Superset version

6.1.0

### Python version

Not applicable

### Node version

I don't know

### Browser

Chrome

### Additional context

_No response_

### Checklist

- [x] I have searched Superset docs and Slack and didn't find a solution to my problem.
- [x] I have searched the GitHub issue tracker and didn't find a similar bug report.
- [x] I have checked Superset's logs for errors and if I found a relevant Python stacktrace, I included it here as text in the "additional context" section.

Start by reproducing the date shift through SQL Lab and a table chart on the reported Windows timezone, then compare the frontend chart rendering with the backend changes in PR 42329. Done means dates remain unchanged in the chart without dataset hour offsets or disabling the temporal-column setting.

Written by the indexing model from the issue text.