Implement ID/LP Parsing

[SamyaDaleh]

G = <V_N, V_T, S R_ID, R_LP>
ID-Rule S -> a, b, c
with LP-Rule b < c
is equivalent to three CFG rules: S -> abc, S -> bca, S -> bac

Implementation is based in Earley, but any symbol after the dot can match the next input symbol. For nonterminals any next symbol can be expanded (that is permitted by all LP restrictions).

Another exmple:
S -> A, S, B
S -> A, B
A -> a, a
B -> b, b
A < B

S. M. Shieber interprets the right side of the dot as multi-set, E. G. Barton interprets both sides as multi-sets. Shieber keeps linearization visible while Barton generates less items. Maybe implement both.