Remove all remaining uses of TypeIds as instruction operands. (#5280)

In preparation for shifting from `TypeId`s potentially representing
attached types to always representing unattached types, using
[terminology suggested on
Discord](https://discord.com/channels/655572317891461132/963846118964350976/1359286326779973712).
This change causes us to track slightly more type spelling information
through SemIR.

One change that has significant impact on the SemIR output is that we
now build a `struct_type` instruction in each class representing the
types of the fields, including the spelling used for those types. This
is now no longer always identical to the corresponding canonical
`struct_type` for the object representation, so it's built separately
and owned by the class.

Also remove `TypeBlock` support entirely, as its only use was
representing `TupleType`s, which now use an `InstBlock`.

toolchain/check/import_cpp.cpp

@@ -524,7 +524,7 @@ static auto ImportCXXRecordDecl(Context& context, SemIR::LocId loc_id,
   // The class type is now fully defined. Compute its object representation.
   ComputeClassObjectRepr(context,
                          // TODO: Consider having a proper location here.
-                         Parse::NodeId::None, class_id,
+                         Parse::ClassDefinitionId::None, class_id,
                          // TODO: Set fields.
                          /*field_decls=*/{},
                          // TODO: Set vtable.