Distributed Backtransform Tridiagonal - Check intra-epoch order performance

[albestro]

An implementation detail of the distributed back transform tridiagonal algorithm has been chosen without performing any check.

Operations are already scheduled scanning the matrix with a smart ordering by so called "epochs", but it has been deliberately imposed an order for the intra-epoch loop.

We don't expect this cause any impact, but it is better to check that we are not missing anything that could be relevant performance-wise.

Originally posted by @albestro in https://github.com/eth-cscs/DLA-Future/pull/633#discussion_r980214302