Minor changes to all models to check

[colemonnahan]

So in doing some testing we discovered a few potential issues that should be checked for all models:

1.) Make sure that years specified for SD reports are updated in the starter.ss file to the new range (1:100). This will only throw an error in some cases.
2.) Change the lbin method in the .dat file to 2 so we're using data length bins for CAL data instead of populations bins. 2 #_Lbin_method: 1=poplenbins; 2=datalenbins; 3=lengths. This is right before the agecomps, NOT before the length comps.

[kellijohnson-NOAA]

Standardization task list:

• lbin method
• starter.ss: use make_starter
• parameter bounds: use standardize_bounds
• fleet names
• forecast.ss: use make_forecast

Please add other tasks that I forgot.

[colemonnahan]

• Priors should be turned off in both models for all parameters (set prior_type=-1)
• Play with phases to ensure good/fast convergence. E.g., if high recruitment devs, estimate those in phase 1 and then R0. Optimal phases depend on the model.
• Let's strip down the .dat file to have minimal data, say 1 year/fleet for each data type. The first OM run will run faster this way. clean_data is run after this so everything gets removed dynamically anyway.
• Set dummy index values for fleets to be used for CPUE for all years. The change_data function could do this but currently doesn't. It's assumed the index data is there.

[seananderson]

• make sure data files all have 100 years of catch (looks like only cod-age still doesn't have 100)
• set Lbin_method to 3 3 #_Lbin_method: 1=poplenbins; 2=datalenbins; 3=lengths Note the capital L. This is right before the agecomps, not before the length comps.

[kellijohnson-NOAA]

I need a minimum age for summary biomass for all species. Should I trust the age in the current starter files or set a default age of 0 or 1 for all species?