Port nocapd to Deno and rename [bullmq->pqueue, lmdb -> sqlite]

[dskvr]

There's a bug on production agents that relate to LMDB silently failing during update once the db reaches a certain size. This results in a diminishing reporting rate, at about 5 relays a day. (for example, in cursory tests I found that the current number of (known) relays online using current logic is approximately 1040, but most monitors are reporting 900 relays, where some monitors are reporting relays online that others aren't. It has been confirmed that this is not due to network conditions, as isolated tests the hosts can indeed connect to the relays they are not reporting.

Fed up with Node, as well as the silent peculiarities of LMDB and BullMQ, I am abandoning everything that brings me dispair.

• node -> deno
• bullmq -> pqueue
• lmdb -> sqlite

There are some increased overhead and performances issues with this change, particularly with sqlite (roughly 1000x slower).

But overall, will be easier to maintain and more reliable, and becomes a stability update, quality of life update (for me) and much needed code refactor that passively kills all technical debt.

Already have it about 90% of the way there, just need to reenter bundler hell because @nostrwatch/nocap relies on ws and Deno and ws has all kinds of unhandled errors in Deno.

[netlify]

❌ Deploy Preview for nostrwatch failed. Why did it fail? →

Name	Link
🔨 Latest commit	da525c3
🔍 Latest deploy log	https://app.netlify.com/sites/nostrwatch/deploys/67aac7edb534e800080a2725