Why does the RawTask use raw pointers

Pure for learning purposes, why does `RawTask` use raw pointers for `output`, `header`, etc? Is it mostly for performance reason?

```
/// Raw pointers to the fields inside a task.
pub(crate) struct RawTask {
/// The task header.
pub(crate) header: *const Header,

/// The schedule function.
pub(crate) schedule: *const S,

/// The tag inside the task.
pub(crate) tag: *mut T,

/// The future.
pub(crate) future: *mut F,

/// The output of the future.
pub(crate) output: *mut R,
}
```

Start with the RawTask definition shown in the issue and trace how its header, schedule, tag, future, and output pointers are created and accessed. Review the surrounding task implementation to identify the ownership and lifetime constraints; the issue is done when those choices are documented clearly, including whether performance is a stated motivation.

Written by the indexing model from the issue text.