no burning at lowT, lower the default jina rate floor.

See : https://lists.mesastar.org/pipermail/mesa-users/2024-May/015169.html

See also https://docs.mesastar.org/en/release-r24.03.1/reference/star_job.html#jina-reaclib-min-t9

"set jina reaclib rates to zero for T9 <= this. if this control is <= 0,
then use the standard default from rates. need <= 3d-3 for pre-ms li7
burning if change this, must remove old cached rates from
data/rates_data/cache"

We currently have `reaclib_min_T9 = 1d-2` (i.e. no reactions below 10 MK) as a MESA default, which misses be9 burning as well as li7 burning on the pre-ms, so these rates are effectively set to 0. This seems like an issue. Why don't we lower this to 1d-4 or something similar to see if it breaks anything? I think we should find a way to lower the floor if possible or communicate when it should and shouldn't be lowered more clearly to users. I was not aware of this until Masanobu raised it and it directly influences pre-ms modeling, but it could have unforeseen consequence we are not yet aware of.

Start with the jina-reaclib-min-t9 reference documentation and the reaclib_min_T9 control described in the issue. Investigate the effects of lowering the floor, including the required data/rates_data/cache handling. Done means a tested decision on the default and clear guidance about when users should lower it.

Written by the indexing model from the issue text.