Stabilization and equal-order elements

[jcallaham]

Comparison of original IPCS solver to BDF3 with GLS stabilization and linear velocity elements

	Cylinder	Pinball	Cavity	Step
IPCS (nproc=1)	0:57	3:34
IPCS (nproc=4)	0:48	1:50
IPCS (nproc=8)	0:43	1:30
IPCS (nproc=32)	1:15	1:43	1:10:30	13:13
BDF3 (nproc=1)	0:08	0:24
BDF3 (nproc=4)	0:08	0:17
BDF3 (nproc=8)	0:09	0:15
BDF3 (nproc=16)	0:10	0:16	2:00	1:05
Max speedup	5.3x	6x	24.9x	12.2x

TODO:

• Rebase after BDF/EXT time-stepper #134 is merged (Failed miserably at this)
• Add stabilization to Newton solver for steady-state
• Move stabilization definitions to their own classes
• Allow for setting Krylov solver via BDF solver options (Schur complement PC seems to be best in all cases)
• How should the environment configuration work? Should we just default to the most efficient settings?
• Consolidate actuator & BFS noise updates and move to flow (out of scope - see Consolidate actuator & BFS noise updates and move to flow #138)
• Settle on best constant definition (alfi?)
• Re-run benchmarks

[cl126162]

Checked the SUPG changes and benchmarked it vs IPCS and the BDF. It is more stable compared to the other solvers and works fine.