Not needed anymore.  We just use operator= now.

This allows partial-instantiating custom Null object for template Lookup<T>.
Before, this had to be handcoded per instantiation.  Apparently I missed
adding one for AAT::ankr.lookupTable, so it was getting the wrong (generic)
null for Lookup object, which is wrong and unsafe.

Fixes https://bugs.chromium.org/p/chromium/issues/detail?id=944346

We'll see if bots like.

Nothing an extra template class wouldn't fix...

Fixes https://github.com/harfbuzz/harfbuzz/issues/1419

This concludes null-size enforcement changes

../../src/hb-null.hh:53:39: warning: enum constant in boolean context [-Wint-in-bool-context]

Fixes https://github.com/harfbuzz/harfbuzz/issues/1300

This reverts commit 442a72d9.

Doesn't make sense.  No idea how my local compilers where happy with it!

Towards fixing https://github.com/harfbuzz/harfbuzz/issues/1146

Fixes https://github.com/harfbuzz/harfbuzz/issues/1299

Removes anomaly I was seeing in cpal table trying to use implicit Null(NameID).

This speeds up Roboto shaping by ~10%.  I was hoping for more.
Still, good defense against lookups with many subtables.

Implemented path operators in a new struct PathProc hooked up to CSOpSet
Added current point to CSInterpEnv

Some clang versions define static_assert as a macro apparently, so we cannot
redefine it...

This reverts commit 94bfea0c.
This reverts commit 4e626278.

So we don't accidentally break it again.

Sorry for the noise, downstream custom builders.  Please adjust.

Finally, unified!