[Bug]: add warning for age 0 maturity

[Rick-Methot-NOAA]

Describe the solution you would like.

A user with an EWAA model had mature fish occurring at age 0.
It was annual model, with spawn_seas = 1.0 and settlement also at month 1.0.
During the time series, the age 0 fish do not get created until after SSB is calculated, so the SSB does not include any age 0 fish even though they are mature.
During the forecast, the multiple loops cause the age 0 fish to get populated in the first loop and then exist in later loops, so get included in the SSB.
This needs to be prevented.

Note that empirical wtatage takes precedence over First_Mature _Age, so adding a warning specific to EWAA

@alexjensen-NOAA

• New warning added for wtatage.ss scenarios:

Warning 1 : WTATAGE.SS has positive fecundity at age 0 for year = 1970 Are you sure? fecundity is only used at the time of spawning and age 0 fish normally do not exist at time of spawning`

• add code to zero out the age 0 fish in the forecast. This is not an issue during the time series.

  natage(t, p, g, 0) = 0.0; // these fish should not exist at beginning of year.  They are created after spawning.  Are here in array only due to Fcast_Loop1
 SSB_pop_gp(y, p, GP4(g)) += fracfemale_mult * fec(g) * natage(t, p, g); // accumulates SSB by area and by growthpattern

• add warning regarding age 0 maturity

  if (First_Mature_Age == 0)  {
    warnstream<<"First_Mature_Age read as:  " << First_Mature_Age << ", which is unusual. Check logic of spawn_month and settlement time & age";

Describe alternatives you have considered

none obvious

Statistical validity, if applicable

NA

Describe if this is needed for a management application

yes, upcoming assessment

Additional context

No response

[Rick-Methot-NOAA]

I added an extra write statement to test what is happening with this setup. The N at age 0 is always 0.0 during the time series when SSB is calculated, but during forecast, it is 0.0 in the first ABC_loop, then populated during that loop and that value gets used for the SSB calc for ABCloops 2 and 3. Robust logic to zero this out in each ABC_loop will be tricky because of possibility of multiple seasons and multiple settlement events.

If spawnseason was set to month 2 and recruitment also to month 2, there already is this error message (even without maturity at age 0.0):

Warning 1 Fatal Error! Cannot have spawn_time_seas after beginning of a season and settlements in the same season
++ put spawning at beginning of the season, or move settlements to next season
Also see warning.sso
Exiting SS3!

So, I will add a similar warning with month 1 if age 0 fish are mature. Unfortunately, the existing warning happens in readcontrol, but SS3 does not know at that time if maturity at age 0 is positive or not. So new warning will need to be in prelim calcs.

@iantaylor-NOAA bringing Ian into the discussion

[Rick-Methot-NOAA]

The existing code logic was for empirical to override first_mature_age per:
"First Mature Age: all ages below the first mature age will have maturity set to zero. This value is overridden if maturity option is 3 or 4 or if empirical weight-at-age (wtatage.ss) is used, but still must exist here."

Although I originally thought to reverse that, instead I will only create a warning note if positive fecundity values are read empirically for ages < first_mature_age