Add feature: dynamic load balancing

[IshaanDesai]

Most important points regarding the implementation:

• When the load balancing is triggered, on each rank, the average active simulations per rank are computed (MPI allgather).
• If a rank has lesser active simulations than the floor value of the global average, the rank needs to receive simulations to make up for the shortage.
• If a rank has more active simulations than the ceiling value of the global average, the ranks needs to send the excess simulations away.

Checklist:

• I made sure that the CI passed before I ask for a review.
• I added a summary of the changes (compared to the last release) in the CHANGELOG.md.
• If necessary, I made changes to the documentation and/or added new content.
• I will remember to squash-and-merge, providing a useful summary of the changes of this PR.