[edkerk]

read_yaml_model reads the ecModel files that GECKO and RAVEN write, and fseof follows GECKO's ecFSEOF where it differs.

YAML reader. read_yaml_model reads the same files as RAVEN's readYAMLmodel, which is line based:

• An ec reaction written as - enzymes: !!omap with nothing under the tag (GECKO and RAVEN write this for a reaction without enzymes) is read as a reaction with no enzymes. It raised AttributeError: 'ScalarNode' object has no attribute 'flow_style'. The tutorial light Human-GEM ecModel.yml has 52 such reactions and eciYali5-GEM has 29.
• A double-quoted value with quotes inside, such as ""text"" or "a "b" c", is read as "text" and a "b" c: the outer pair of quotes is dropped and \\ and \" are unescaped. Older RAVEN (writeYAMLmodel with preserveQuotes) wrote a note that carried its own quotes this way, and eciYali5-GEM has three of them.
• A single-quoted value with an apostrophe inside, such as 'it's', is read as it's.
• The repairs run only when the strict parse fails, so files that parse as YAML are read unchanged. A comment after a repaired value is not handled.

fseof. fseof solves each level the way RAVEN's solveLP and GECKO's ecFSEOF do:

• The pFBA step holds the objective at 0.999999 of its optimum at that level. It held the full value, and on the scipy solver the optimum is only accurate to HiGHS's feasibility tolerance, so the step was infeasible at the top level of ecYeastGEM and the scan returned 9 of 10 levels. It returns all 10, as on Gurobi and GLPK. A level at which the biomass maximisation is infeasible still ends the scan.
• The top of the scan is max_fraction times the target flux in the pFBA solution that maximises the target, where it took the plain maximum (a difference of about 1e-6 relative).
• Targets of equal score are listed in model order (a stable sort).
• The new option equality=True fixes the target flux at each level, as ecFSEOF does. The default keeps the lower bound of RAVEN's FSEOF.

Used by CompMetEng/geckopy#72 and the comparison in CompMetEng/raven-gecko-parity#141.