Add a custom html dataframe rendering view of nested columns

[dougbrn]

We have control over the dataframe output within Pandas: https://pandas.pydata.org/docs/user_guide/style.html

When left unstyled, Pandas tries to print the dataframe directly into the cell values. We should have a custom rendering that changes this to be some simple representative value, like: <NestedFrame>. Optionally we could try to give this some more descriptive metadata, but I worry about the overhead of calculating such information if the user doesn't need it. e.g. <NestedFrame (length=350)>

[dougbrn]

This code is likely what we're after, it alters the default representation without converting it to a styler object:

def my_style(df:NestedFrame):
    return (df.style.format({"nested": ""}))

def _my_repr_html_(self) -> str | None:
    return self.pipe(my_style)._repr_html_() # <<<< !!! HERE !!! 
        
NestedFrame._repr_html_ = _my_repr_html_

Note that one quirk of this is that my_style needs to know what columns to style manually, so we may need to trigger this in add_nested calls and reassert the styling.

[hombit]

Connected to #64