Assignment-arc cost is added twice to the netflow objective

Problem

In ets_fiber_assigner/netflow.py, buildProblem() accumulates the per-arc
cost into the objective twice (lines 513 and 517):

tcost = vis_cost[ivis]
if cobraMoveCost is not None:
    dist = np.abs(bench.cobras.centers[cidx]-tpos[ivis][tidx])
    tcost += cobraMoveCost(dist)
prob.cost += f*tcost                                    # line 513
if blackDotPenalty is not None:
    dist = np.min(np.abs(closestDotsList[cidx]-tpos[ivis][tidx]))
    tcost += blackDotPenalty(dist)
prob.cost += f*tcost                                    # line 517

Because tcost accumulates, the visibility and cobra-move terms are charged
twice for every Tv_Cv arc. With blackDotPenalty configured, an arc
contributes:

actual:   2*(vis_cost + cobraMoveCost) + blackDotPenalty
intended:   (vis_cost + cobraMoveCost) + blackDotPenalty

Looking at the commit history, this seems a copy-paste leftover rather than deliberate weighting.

Impact

Affects every backend (Gurobi, HiGHS, PuLP). The visibility and cobra-move terms
carry double their configured weight relative to nonObservationCost,
partialObservationCost and blackDotPenalty, so the optimizer trades them
off differently than the configuration intends.

Proposed fix

Remove the first accumulation (line 513).

Note for review

This changes results wherever vis_cost or cobraMoveCost is nonzero:
correcting it effectively halves those terms relative to the other costs, and
fiber assignments will shift. If the cost functions were tuned empirically
against the current behavior, some caution has to be observed, or rescaling in the netflow would be preferable.