When a context menu is open and the user right-clicks a different item,
the overlay's close handler resets the menu state but no longer captures
the event (since e10459f). The right-click propagates to the widget,
which reopens the menu with stale `MenuBounds` from the previous context
menu. If the new menu has a different number of items, this causes an
assertion failure in `MenuState::layout`.
Call `data.reset()` before setting `data.open = true` in the right-click
handler to ensure stale `MenuBounds` are cleared before the next layout
pass rebuilds them.
This also changes `widget::column::with_children` and
`widget::row::with_children` to take an `impl IntoIterator` instead
of a `Vec`, like the `iced` variants of these functions do.
This shouldn't be a breaking change since passing in a `Vec` will still
compile and function exactly as before.
(Using `iced::widget::Column::from_vec` or
`iced::widget::Row::from_vec` isn't possible, since the elements of the
`Vec` aren't checked, so the size of the resulting `Column` or `Row`
won't adapt to the size of its children. Perhaps a new function could
be added to mirror `iced`'s?)
To reduce compile-times and avoid some overhead to binary size, this will modify some of our
generic functions to use non-generic inner functions where possible. The inner functions are
marked carefully with `#[inline(never)]` to prevent being inlined by LLVM at their callsites
While looking for generic functions to optimize, I have also taken the opportunity to annotate
public non-generic getters and setters with `#[inline]` to ensure that LLVM will inline them
across crate boundaries. By default, only generic functions are automatically inlined, and
only when enabling fat LTO are constant functions reliably inlined across crate boundaries.
