Clean up print statements and maybe use logging package

[rickecon]

We should explore cleaning up what gets printing during an OG-Core run and how it gets printed.

1. Currently, the steady-state GE loop errors show up as np.float64(). We should format these output to show fewer decimals and to not show the np.float.
2. Copilot gave a [nitpick] suggestion in PR Add comprehensive Dask performance benchmark tests #1049 to "Consider using logging instead of print statements for better control over output in test environments. This would allow filtering of debug output during test runs." Change line 186 in test_real_txfunc_benchmarks.py from print(f"Generated {len(data)} valid tax records for year {year}")tologging.info(f"Generated {len(data)} valid tax records for year {year}")`.
3. Revisit the other output during a compete run and decide what is necessary and desirable.

cc: @jdebacker

[jdebacker]

@rickecon (2) was mostly implemented in PR #1005 (the motivation for this was that I didn't want all the output printed when trying to solve the model many times as part of a SMM estimation). I completed this in PR #1049 (remaining print statements are only in scripts that are not necessary to solve the model (e.g., execute.py and run_ogcore_example.py) and therefore don't result in repeated print commands when the model is run.

I agree on solving (1) and an open to discussing (3).

[jdebacker]

Remaining issues resolved in PR #1053