As 503748d8 fix

also snake_cased popSubrtNum and other surrounding function names

Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=14641

The iter objects shouldn't not be const.  D'oh.

The bool conversion happens after predicate is called automatically.

WHy do only some of the clang bots catch this I have no idea :(.

Make it an error.

Let's see if fixes MSVC fail.  Though, the error doesn't make sense to me.

  hb-blob.cc
c:\projects\harfbuzz\src\hb-algs.hh(166): error C2440: 'return': cannot convert from 'unsigned int' to 'unsigned int &&' [C:\projects\harfbuzz\build\harfbuzz.vcxproj]
  c:\projects\harfbuzz\src\hb-algs.hh(166): note: You cannot bind an lvalue to an rvalue reference
  c:\projects\harfbuzz\src\hb-algs.hh(174): note: see reference to function template instantiation 'T &&<unnamed-type-hb_min>::impl<T,unsigned int&>(T &&,T2) const' being compiled
          with
          [
              T=unsigned int,
              T2=unsigned int &
          ]

Allows for eg, checking all values equal 2: hb_all (it, 2).

They do absolutely same thing.  hb_requires is to encode constraints,
whereas hb_enable_if is for more conditional enabling.