We should run a mini crawl frequently

• neume-network/data runs a full crawl in roughly 2-3hrs every day
• But to get e.g. feedback whether some strategies are properly integrated with each other can hence take a very long time. E.g. if we ship a change in strategy X, and downstream strategies break because of it, it may take up to one day for this to surface in a crash of the neume-network/data crawl
• However, we don't have to crawl the entire chain and all platforms to get the information whether e.g. all strategies are properly integrated with each other. Instead, we could e.g. set up web3subgraph or any other starting strategy to e.g. only return a single id e.g. [{"id":"0x01ab7d30525e4f3010af27a003180463a6c811a6/1","createdAtBlockNumber":"14323199","platform":"sound"} and then do the crawl for just that id. (I've just taken the first value of: https://raw.githubusercontent.com/neume-network/data/main/results/web3subgraph-extraction)
• If we found a set of start ids that resulted in all strategies being invoked at least once, we could probably finish an entire mini crawl in a matter of seconds or minutes: and we'd have a fast response/feedback for making coding mistakes.

Start by reviewing the full crawl in neume-network/data and the example value from results/web3subgraph-extraction. Identify a representative set of starting IDs that invokes every strategy, then determine how a mini crawl could run frequently. Done means the reduced crawl completes quickly and provides feedback when strategy integrations break.