Cleaner log for adjusted splits

[Nexesenex]

With this PR, the logging now looks like this :

================================ max_gpu = 2
Adjusted split at layer  0:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted split at layer  8:  GPU0:    0 ; GPU1:  0.5 ; GPU2:  0.5
Adjusted split at layer 16:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted split at layer 24:  GPU0:  0.5 ; GPU1:    0 ; GPU2:  0.5
Adjusted split at layer 32:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted split at layer 40:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted splits (totalized): GPU0:    0 ; GPU1:    0 ; GPU2:    0

All aligned cleanly in the loading screen, with a totalizator. (Edit: I need to fix that one lol..)

Ok, fixed. ^^

Adjusted split at layer  0:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted split at layer  8:  GPU0:    0 ; GPU1:  0.5 ; GPU2:  0.5
Adjusted split at layer 16:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted split at layer 24:  GPU0:  0.5 ; GPU1:    0 ; GPU2:  0.5
Adjusted split at layer 32:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted split at layer 40:  GPU0:  0.5 ; GPU1:  0.5 ; GPU2:    0
Adjusted splits (totalized): GPU0:  2.5 ; GPU1:  2.5 ; GPU2:    1

Also, I added 2 forgotten items in the --minilog blacklist of the sweep-bench.

• I have read the contributing guidelines
• Self-reported review complexity:
  • Low
  • Medium
  • High

[Nexesenex]

Sorry for the mess @ikawrakow. I was editing as you were reviewing.
Now, it compiles and works properly, addition included.

[magikRUKKOLA]

@Nexesenex

The best way to display such a data would be to write the divisor of 1 (the number of GPUs used in the split) and the positions. So:

  Adjusted split at layer  0:   0  0  0  0  0  0  2  0  0  2
  Adjusted split at layer 12:   0  0  0  2  2  0  0  0  0  0
  Adjusted split at layer 24:   2  2  0  0  0  0  0  0  0  0
  Adjusted split at layer 36:   0  0  2  0  0  0  0  0  2  0
  Adjusted split at layer 48:   0  0  0  0  0  2  0  2  0  0

*extra space for the cases of setups of more than 10 GPUs and even distribution of splits across all of the GPUs.

This way there is no need to use 0.5 anymore (which is confusing since the 0 are also used to signify the absense of the work for a certain GPU.