determine z range for localization stacks from widefield stacks and queue them - #906

posting here for discussion - one of the three options suggested in #898. "works", but with a Hoechst stain works more for finding the center of focus rather than the bounds (not robust enough for production runs at the moment and I need to tack back to other priorities).

Proposed changes:

• allow stack setting control in startspool (separately PR'd more seriously in allow stack settings to be per-series #902)
• hack a fix around [BUG] PYME acquisitions lacking a z-controller try to access it nevertheless #714, but incomplete - didn't commit that dwellTime = None needs to be 1 or so
• recipe module to determine stack settings from widefield stack (uses fixed lower bound to determine metric threshold) and add it to len-1 tabular metadata so it can be used in:
• recipe module to post spool series action, pipping in stacksetting metadata from input metadata