Run allocation at t = 0

[SouthEndMusic]

Fixes #1389.

[SouthEndMusic]

I am struggling a bit to understand the behavior of the allocation_example model at the start of the simulation. Here is the plot of the allocation data:

And here is the data for the first allocation solve:

 Row │ time     subnetwork_id  node_type   node_id  priority  demand   allocated  realized 
     │ Float64  Int32          String      Int32    Int32     Float64  Float64    Float64  
─────┼─────────────────────────────────────────────────────────────────────────────────────
   1 │     0.0              2  UserDemand        6         1      0.0        0.0       1.0
   2 │     0.0              2  UserDemand        3         1      1.5        1.5       1.5
   3 │     0.0              2  UserDemand       13         1      0.0       -0.0       1.2
   4 │     0.0              2  UserDemand        6         2      0.0        0.0       1.0
   5 │     0.0              2  UserDemand        3         2      0.0       -0.0       1.5
   6 │     0.0              2  UserDemand       13         2      1.2        0.2       1.2
   7 │     0.0              2  UserDemand        6         3      1.0       -0.0       1.0
   8 │     0.0              2  UserDemand        3         3      0.0       -0.0       1.5
   9 │     0.0              2  UserDemand       13         3      0.0       -0.0       1.2

The abstracted a.k.a. realized values at $t = 0$ can be ignored, as there is no realized flow at $t = 0$ yet. These are the things I find weird:

• UserDemand node 13 having a positive demand of 1.2 at priority 2. In the schematization this node is only assigned a demand at priority 3. Things are getting mixed up somewhere. This also causes nothing to be allocated to UserDemand node 3 at priority 2;
• UserDemand node 3 having 0 abstraction after the first day, even though flow was allocated to this node.

Edit: I am more confused still. I noticed that some flow goes to the buffer for the fractional flow even though it is not needed there. There might also be a difference between the allocation_example model from the docs and the one in the test models.

[visr]

Nice! Left a few comments.