Revert "Update opaques.md"

[mbloms]

Reverts #10976

This was not a typo.

The Access object defines three opaque type aliases:

• Permission, representing a single permission,
• Permissions, representing a set of permissions with the meaning "all of these permissions granted",
• PermissionChoice, representing a set of permissions with the meaning "at least one of these permissions granted".

permissionSetX & permissionSetY is supposed to be the union of permissionSetX and permissionSetY, not the intersection of them.

It's also clear that def & (y: Permissions): Permissions = x | y was intentional from this text:

Also, the definition of ReadWrite must use |, even though an equivalent definition outside Access would use &.

What is meant by this is that the real method & would take precedent over the extension method &, so | must be called manually instead of & in the Access object.

The idea with this example is to show that it's useful that a single permission has the same representation as the equivalent singleton Permissions set or the equivalent PermissionChoice set.

If one wanted to, the subtyping relation could be changed to this:

//"All permissions in X granted" implies "At least one permission in X granted"
opaque type Permissions <: PermissionsChoice = Int
//At least one of these permissions granted
opaque type PermissionChoice = Int
//"A single permission granted" implies "All permissions in this singleton set granted"
opaque type Permission <: Permissions = Int

The example demos how opaque types behaves in many ways at the same time, perhaps even too much in too little text, hence why it's a bit confusing.

[b-studios]

My bad, when reviewing the PR I only looked at the changes without looking at the text so got it wrong as well.