GAMG: support for polling

Summary

processorGAMGInterfaceField does not implement ready() even though it maintains its own requests. Hence any polling will call matrixInterfaceUpdate in original order, negating any benefit of polling. Note that it only affects operations inside the GAMG solver.

Steps to reproduce

Hard. Default lduInterfaceField::ready() returns true so it will do the waiting instead in the matrixInterfaceUpdate routine.

What is the current bug behaviour?

Order of GAMG interface updates is always the same.

What is the expected correct behavior?

Re-runs might give slightly different truncation error.

Relevant logs and/or images

Environment information

• OpenFOAM version : v2309