For MIRI use first group when bright

[stscijgbot-jp]

Issue JP-3799 was created on JIRA by Karl Gordon:

For MIRI, the first group (aka frame) is flagged as DO_NOT_USE due to the firstframe effect. In the case of a pixel having a large signal, it has been found that the slope estimated from the 2nd-1st groups and 3rd-2nd groups is equivalent within 1% indicating that the firstframe effect is small compared to the signal in this specific case. This means that for pixels seeing very bright signal (i.e., saturating between the 2nd and 3rd groups), that the slope can be measured from the 1st and 2nd groups. This provides a slope measurement where previously there was none (i.e., a NaN value). To enable this in the pipeline, the firstframe flagging for these specific pixels needs to be disabled.

A implementation of this for the jwst firstframe step exists. This behavior (not flagging for very bright signals) is controlled by a keyword that is set to False by default.

A report on the analysis that lead to this finding is ready and will be undergoing review soon.

[stscijgbot-jp]

Comment by Melanie Clarke on JIRA:

Implemented in #8952