ph5_validate not seeing all post pickup windows (nodes only)

[hrotman-pic]

Example experiments: 22-004 (Fairfield), 22-025 (SmartSolo). Both examples are single deployment experiments (no moving of instruments to a different station). These cases were found during PH5 to TileDB experiment analysis review. For 22-004, the Das_t information is from the PH5 to TileDB analysis detail csv generated by Matt Briggs, which is obtained from the Das_t in the DMC experiment holdings. For 22-025, the Das_t information is looked up via kefedit in the Das_t in the experiment at EPIC. To my knowledge, they are equally accurate methods for seeing the extent of a given Das_t.

22-004 example:
Station 3011 has pickup time 23:39:00 on January 13 in Array_t. The Das_t for this station (1X3011) continues until approximately 07:00:00 on January 15. The ph5_validate warning for this station is:
WARNING: Data exists after pickup time: 1352 seconds.

22-025 example:
Station 9074 was not turned off when intended. The Array_t pickup time is 17:00:00 on August 20. The Das_t for this station (9X9074) continues until approximately 16:00:00 on August 25. The ph5_validate warning for this station is:
WARNING: Data exists after pickup time: 1800 seconds.

This could be considered a bug, but the check is only a warning. I think this means that if pickup times are adjusted manually in Array_t, the user must be careful since these cases suggest ph5_validate will not show if a large chunk of data is accidentally orphaned. If there is a conflict between addressing this and retaining ph5_validate's ability to handle cases where a Texan is rolled to different stations, I recommend the latter take priority.

Does this affect pre-deploy as well? Unknown, we may need to search for a case with data more than 1 recording window in length in order to determine this.