fix: use bad peer removal logic only for lightpush and filter

[chaitanyaprem]

Description

#1241 had some side-effects where-in store nodes were being removed from peerstore which we do not want because in status store nodes are fixed (i.e fleet nodes).

• Apart from that i had noticed that when node is offline, connections are being triggered which result in failures and many good peers such as fleet nodes are getting removed from peer-store.
• Added a more strict check and now migrated the bad peer removal logic to only be done from lightpush and filter and that too based on specific errors to avoid such side-effects.
• Found an issue with offline to online state change handling in filter-manager where filter ping to existing subs will never happen.
• Added a check to filter ping error handling in case node is already offline so that sub are not cleared

This still needs to be tested, will update the PR as i get more info from the testing.

[status-im-auto]

Jenkins Builds

Click to see older builds (1)

❔	Commit	#️⃣	Finished (UTC)	Duration	Platform	Result
✔️	5b86d20	#1	2024-10-29 11:02:30	~7 min	unknown	📄log

❔	Commit	#️⃣	Finished (UTC)	Duration	Platform	Result
✔️	371aa87	#2	2024-10-29 11:13:18	~2 min	unknown	📄log
✔️	371aa87	#3	2024-11-04 07:18:12	~2 min	unknown	📄log
✔️	371aa87	#4	2024-11-04 07:18:32	~13 sec	unknown	📄log
✔️	b646eff	#5	2024-11-05 17:43:04	~2 min	unknown	📄log

[chaitanyaprem]

Closing this as changes are being considered in #1244