I generalized one of the assertions a little bit and merged this.
I verified that it fixes the pegsol problem; in Sokoban, I couldn't actually
reproduce the original problem.

Got rid of the assertions.
Pushed into downward-fixes.

The second error is also a wrong assertion.
The assertion asserts (:-) that a search node can be declared as a dead-end only 
when it's new (i.e., the first time it's evaluated).
However, the landmark heuristics as well as the stuff from my HDIP paper in 
Freiburg are path-dependent, and so the assertion is not true for them.

I will get rid of it.

Does the assertion serve an important semantic point?
If not, it may make sense to get rid of it.

The first error 
(landmarks/landmark_count_heuristic.cc:117: int 
LandmarkCountHeuristic::get_heuristic_value(const State&): Assertion 
`all_costs_are_zero' failed.) 
is due to an assertion which is wrong for the admissible heuristic.
The assertion checks that if the heuristic value is 0, then the min_cost of 
every goal landmark is 0. However, min_cost is a property that is only used by 
the inadmissible heuristic, and corresponds to the minimum cost of all the 
_first_ achievers of the landmark. It's possible that the landmark is needed 
again, in which case we would need the minimum of _all_ achievers.

I can either correct the assertion or get rid of it - any preferences?

I'm nosy-ing Silvia just in case.

assertion error 1:
downward-1: landmarks/landmark_count_heuristic.cc:117: int 
LandmarkCountHeuristic::get_heuristic_value(const State&): Assertion 
`all_costs_are_zero' failed.

config = '--search 
"astar(selmax([lmcut(),lmcount(lm_merged([lm_hm(m=1),lm_rhw()]),admissible=true)
],training_set=1000),mpd=true)"'

problem: pegsol-08-strips:p02.pddl



assertion error 2:

downward-1: search_space.cc:127: void SearchNode::mark_as_dead_end(): Assertion 
`info.status == SearchNodeInfo::NEW' failed.

config = '--heuristic 
"hlm,hff=lm_ff_syn(lm_rhw(reasonable_orders=true,lm_cost_type=2,cost_type=2))" -
-search "iterated([    lazy_greedy([hff,hlm],preferred
=[hff,hlm]),    lazy_wastar([hff,hlm],preferred=[hff,hlm],w=5),    
lazy_wastar([hff,hlm],preferred=[hff,hlm],w=3),    
lazy_wastar([hff,hlm],preferred=[hff,hlm],w=2),   
 lazy_wastar([hff,hlm],preferred=[hff,hlm],w=1)],    
repeat_last=true,continue_on_fail=true)"'

problem: sokoban-sat08-strips:p15.pddl