The first thing I'd check is if WebLink is regularly crashing due to the issue described in KB 1014527: The May 14, 2024 Cumulative Update for .NET Framework May Cause Reliability Issues with the Laserfiche Web Client and Laserfiche WebLink.
If the WebLink IIS Application Pool w3wp.exe process is crashing (and then rapidly restarting), users would have to log back in because their session would be for the old process that crashed, not the new one. In this scenario, the sessions aren't "timing out", they're being terminated.
No timeout-related config setting will make any difference if that's what's happening, so you'll want to either address that issue or rule it out before proceeding down other paths. The KB article linked above includes instructions on how to identify and mitigate the crash issue.
Edit: You could also check in the Laserfiche Administration Console if a global repository inactivity timeout is set, though I suspect that's not the case as it would affect your Web Client and Windows Client users as well. See: Repository Administration > Repository Options > Idle Disconnect and Last Login