Add known issue with restart reproducibility when changing CM2 cycle length

[anton-seaice]

As noted here and confirmed by @MartinDix, changing the cycle length in CM2 can give different answers. This is a failure in restart reproducibility.

For example, two runs of 1 year length with a restart in between will give slightly different results to a single run of 2 year run length. As such, users need to be aware that for consistency between experiments, they should not change the run length across their experiments.

We are not planning on fixing this in the model, so should add this as a caveat/known issue in the Run ACCESS-CM2 docs.

[access-hive-bot]

This issue has been mentioned on ACCESS Hive Community Forum. There might be relevant details there:

https://forum.access-hive.org.au/t/different-cycling-frequencies-in-access-cm2-lead-to-different-solutions/4539/4

[atteggiani]

We are going to add a warning/known-issue message in the "How to run CM2".