Consider showing UTC offset in tooltip for datetimes

Say I have

import pandas as pd
dfa = pd.DataFrame(
    {
        "datetime": ["2020-10-25T01:30"] * 2,
        "price": [40, 50],
    }
)
dfa['datetime'] = pd.to_datetime(dfa['datetime']).dt.tz_localize('Europe/London', ambiguous=[True, False])
dfa

                   datetime  price
0 2020-10-25 01:30:00+01:00     40
1 2020-10-25 01:30:00+00:00     50

and I plot

px.scatter(dfa, x="datetime", y="price")

then I get

Notice how the two points are stacked one on top of each other, even though they are actually 1 hour apart

If Plotly's design decision for time zones is to display the local datetime, then I think this makes perfect sense, nice one 👍 However, I think the tooltip should give enough information to distinguish the two points, whereas currently, they both show:

datetime=Oct 25, 2020, 01:30

I would expect the bottom one to show datetime=Oct 25, 2020, 01:30 +01:00 and the top one to show datetime=Oct 25, 2020, 01:30 +00:00

Start by running the Python reproduction in the issue and inspect how Plotly formats datetime values in hover tooltips. Done means the two ambiguous Europe/London datetimes remain distinguishable by showing their UTC offsets, with coverage added for this example.