bugfix worse lower bound than incumbent

[Wikunia]

Only add a new solution if it can produce better results than the current incumbent.
Also if the lower bound is worse after evaluating the node we don't want to falsely update the lower bound such that it looks like the lower bound is worse than the incumbent

[codecov]

Codecov Report

Merging #36 (c3d45a9) into main (d226392) will not change coverage.
The diff coverage is 100.00%.

❗ Current head c3d45a9 differs from pull request most recent head 98c2b18. Consider uploading reports for the commit 98c2b18 to get more accurate results

@@            Coverage Diff            @@
##              main       #36   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            4         4           
  Lines          127       128    +1     
=========================================
+ Hits           127       128    +1     

Impacted Files	Coverage Δ
src/Bonobo.jl	100.00% <100.00%> (ø)
src/node.jl	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d226392...98c2b18. Read the comment docs.

[matbesancon]

Only add a new solution if it can produce better results than the current incumbent.

depends, one could want to store multiple solutions as discussed another issue, but the first / last one should always be the best solution yes

[Wikunia]

Here it's more about adding a new node only if it can produce results better than the current incumbent.

[matbesancon]

alright then looks good

[matbesancon]

should we merge this?