StaticHeader::setup presents numeorus opportunities for procedure extraction

[mulkieran]

Basically, all actions taken are symmetric. So, if sigblock 1 read reaches state A and sigblock 2 read reaches state B, then the same action is performed as if sigblock 2 read reached state A and sigblock 1 read reached state B, except that the roles are switched. It makes sense to use one function for these identical actions, rather than two distinct blocks of code.

[mulkieran]

Currently blocked by #1585, but that's a simple one and should go in fast.

[mulkieran]

@GuillaumeGomez Some of the work under this issue was already begun in #1562. If you could, please do this work refactoring separately. It should make it easier to conclude #1562.

[mulkieran]

unblocked now.

[mulkieran]

I think it might make sense to work from the outside inward.

So the outermost 2-by-2 table represents the result of calling StaticHeader::read and looks like:

            Ok        |        Err
Ok  |                 |          x
----------------------------------------------
Err |        x        |

The two quadrants marked x are identical except for switched roles, so one function should do for both. They match up with the blocks starting at 397 and 412.