output of `yarn start:contract` is disorganized, not checked

[dckc]

• excessively verbose
• lacks timestamps, making things like yarn start:contract fails: Proposal rejected at propose-start-contract.sh #44 hard to diagnose
• silent failures: silent failure on insufficient IST when installing bundles #60

For steps like voting, the output should be concise: timestamp, block height, hash, proposal number, YES should fit on one line.

using json-lines uniformly can be a big help.

[dckc]

@amessbee nice work on #27! The logs of the first part of start:contract are so much nicer...

Run yarn start:contract
Waiting for blockchain to start...
..............................................................................................
agd-1  | 2025-01-09T11:03:[4](https://github.com/Agoric/dapp-offer-up/actions/runs/12688514812/job/35365216310#step:8:5)5.545Z block-manager: block 1227 commit
Blockchain is running and producing blocks...

If you have more time/energy for this sort of thing, it would be great to get the rest of the logs cleaned up, as noted in this issue.
Using the dots technique in more places would be cool... we could use a pattern from npm / yarn:

1. put the whole noisy log in a temp file.
   • perhaps prepend a timestamp to each line
2. yarn shows a spinner for progress; we can do a dot for every line of logs, as above
3. If things succeed, just print a summary of the results (last log line? contract name?)
4. If things fail, print something like FAILED to start:contract. see log file /tmp/log234235.
   • Maybe suggest yarn docker:logs as well.

I suppose detecting whether things succeed or fail could be a lot of work; as noted above, the results of several of the steps are just not checked. We could just do here what our ci job does: do a vstorage query for the contract name in published.agoricNames.instance.

[amessbee]

Sounds like a plan. I am going to assign it to myself to get it to when I can without any deadlines in mind if that's okay.