Need to change trials factors for bursts/CBC searches as are currently operating

Following discussion with CBC pipeline leads/group, the trials factor for CBC pipelines should be adjusted to 4 (instead of the currently used 5). This number (4) corresponds to pyCBC, gstlal, SPIIR and MBTA. THe factor of 5 that has been used since the start of O3 was reflecting erroneously as separate pipelines the two versions of SPIIR (low/high mass). These two turned out that are effectively the same pipeline simply reporting under two "search" IDs. As part of this process, CBC pipeline leads confirmed that their reported FARs are indeed "global" FARs over the entire parameter space the searches cover.

At the same time we should reduce to 3 (instead of the currently used 4) the trials factor for online Burst searches. This is in order to reflect the fact that oLIB is not running in the production system and is not producing alerts (its running in the playground is not consistently checked and/or used in real-time).

Trials factor should be updated once RAVEN becomes fully operational, issuing OPAs. When this happens, the trials factors should be cranked up by 1 to 5, for CBC and 4 for Bursts.

Depending how the present running of RAVEN is perceived, one may argue that it is de facto operational, if we may use the assertion of the EM_COINC labels to events to manually trigger OPAs. This step may require manual checking of the joint FARs (at least for GRBs) and manual issuing of GCNs. Depending on how RAVEN reviewers and lead developers feel about the present state of RAVNE, I am amenable to leave trials factors at present level (5 for CBC and 4 for bursts) on the basis of this argument.

Independently, we will need to complete the documentation of such accounting (and potential change) of trials factor in the EMfollow userguide: userguide#94 (closed) .