augment QUDA interface and implement appropriate QUDA fermionic forces

[kostrzewa]

This is the major performance bottleneck left for a reasonably efficient (>70% GPU utilisation) HMC

• argument to set correct format for force in computeCloverForceQuda
• support for opposite parity in computeCloverForceQuda
• support for two flavour fields in computeCloverForceQuda
• think about how to deal with determinant ratios
• cloverdet
• cloverdetratio
• ndcloverrat
• cloverrat
• detratio
• det

[kostrzewa]

For reasons of simplicity it makes sense to run the inversions and get back the solution. This avoids having to duplicate all the logic that we have in place for differentiating between the different types of monomials in QUDA.

For reasons of performance it makes sense to work entirely in QUDA from that point onwards, although there are some complications related to the different types of monomials.