Clean up and add example for C.32 - raw pointers

[bgloyer]

Added a simple example for C.32: If a class has a raw pointer (T*), consider whether it might be owning

Also removed mentioning of owning references. I think owning referecnce are rare and generally not a problem in legacy code.
R.3 gives an exception for legacy code to have owing raw pointers but R.4 does not for owning references. If we want to keep the owing references C.33 should probably be updated to delete owing references.

[cubbimew]

I suspect both LegacyClass and m_owningPtr` run afoul of NL.10

[hsutter]

Editors call: Owning raw pointers and references are already banned under R.3 and R.4. Sometimes they do get used, so we think owning references should still be mentioned. Otherwise, the PR looks good, if you can please restore the owning references we can approve it.

[cubbimew]

How about extending the example to something like

foo* owning = new foo; // change to unique_ptr or owner<T*>
bar* observer = &b; // keep
baz& owning_ref = *new baz; // owning reference; banned
xyz& observer_ref = x; // keep

[bgloyer]

How about extending the example to something like

foo* owning = new foo; // change to unique_ptr or owner<T*>
bar* observer = &b; // keep
baz& owning_ref = *new baz; // owning reference; banned
xyz& observer_ref = x; // keep

Is it enough just to leave examples for pointers? It is true that when working with legacy code, ownership of member pointers is often not clear and requires some digging. Owning member pointers was a common practice in older C++. That is not true for references. Owning references have always been a bad idea. Why would anyone have used an owning T& instead of just T? I'm not sure I have ever seen one in the wild. Even in legacy classes, references should be assumed to be non-owning without second guessing if they are really owned. There are many more owning pointers than references in legacy code - I like examples that reflect what is common.

[bgloyer]

I merge the latest so the diff's are clean now. Anything more on this one?

[hsutter]

Editors call: Thanks!