DOC: revise __astropy_table__'s documentation ? #16235

I noticed that __astropy_table__'s documentation is using a strict_copy boolean argument as an example on an additional keyword argument that can be added on the side of the implementor.
See https://docs.astropy.org/en/stable/table/construct_table.html#table-like-objects

In the context of NumPy's change of semantics regarding the copy argument (see #16167 for the broader discussion), it seems to me that this example will not age well and might become confusing (since the upcoming convention, dictated by numpy, is that copy arguments are already strict). We might also want to update the __astropy_table__ protocol definition so that the meaning of the copy argument aligns with the rest of the code base (i.e., follows numpy at runtime).

In short, I see two documentation changes to be made:

• clarify that the copy argument should follow numpy's semantic at runtime (and demonstrate how)
• use a different example for an additional keyword argument

I feel this needs to be addressed in astropy 6.1.0, though I think (?) it could be done past the feature freeze (scheduled for this Friday). Just in case though, let me ping a couple stake holders here: @astrofrog @saimn @mhvk @taldcroft
To be clear, I can absolutely deal with this before Friday, I'd like to know whether that's required or even desired. Thanks !