Extract process models with pm4py

[abrichr]

Feature request

We would like to convert a recording of one or more workflow iterations into a repeatable and versionable process representation, i.e. a process model.

https://pm4py.fit.fraunhofer.de/static/assets/api/2.7.9/api.html#llm-integration-pm4py-llm

We wish to extract process logs from Screenshot and ActionEvents in order to generate a process log, e.g.

select r.id as case_id, 
	we.title as activity, 
    -- ae."timestamp" as timestamp,	
    datetime(ae."timestamp", 'unixepoch', 'localtime') AS "timestamp",
	COALESCE(ae."timestamp" - LAG(ae."timestamp") OVER (ORDER BY ae."timestamp"), 0) as costs,
	ae.name	as resource
from recording r
inner join action_event ae on r."timestamp" = ae.recording_timestamp 
inner join window_event we on r."timestamp" = we.recording_timestamp and we."timestamp" = ae.window_event_timestamp
where r.id = 1
order by r.id, ae."timestamp";

Replaces #545

Motivation

Extract process logs to generate process model that can serve as replay state machine in model prompt.

[abrichr]

Each process log entry should contain at least the following:

• window title
• action name (e.g. "click", "type", etc.)
• action target (for mouse events this would be the active_segment_description, for keyboard events this would just be the text property)