We now have two files that must be updated for config fuzzer parameters. The mongo_fuzzer_configs.py and now config_fuzzer_limits.py.

This makes configuration complex and increases the risk of mistakes.

I like the idea of having a way to configure limits that isn't in code, but we shouldn't require updates in two files to make this work correctly.