Emit helpful error when db needs to be cleared

[xianny]

Context

When running Mesh with 0xorg/ganache-cli, I sometimes encounter an obscure blockWatcher error. The underlying cause is that Mesh still has the old blocks stored, that are at a higher block height than the snapshot.

It would be nice to emit a helpful error recommending that the user deletes their db folder.

Steps to Reproduce

Please provide detailed steps for reproducing the issue.

It's hard to reproduce, but I first encountered it when restarting ganache without removing the db folder.

This might work:

1. Run mesh + ganache:5.1.0 together (e.g. with docker) and persist the db folder from mesh
2. Stop the processes, update ganache to 6.0.0 and rerun without removing the db folder

Failure Logs

mesh_1_2553ba194d8f | {"error_string":"timed out waiting for first block to be processed by Mesh node. Check your backing Ethereum RPC endpoint","level":"error","msg":"core app exited with error","myPeerID":"16Uiu2HAmN6BkBp9mDZCDt2ptXBVa9Jb3dukTUGa8oumNBaZhNgzx","time":"2020-02-25T21:02:55Z"}

I think this was coming from this section of the code:

// If Mesh is not caught up with the latest block found via Ethereum RPC, ensure orderWatcher
	// has processed at least one recent block before starting the P2P node and completing app start,
	// so that Mesh does not validate any orders at outdated block heights
	isCaughtUp := app.IsCaughtUpToLatestBlock(innerCtx)
	if !isCaughtUp {
		if err := app.orderWatcher.WaitForAtLeastOneBlockToBeProcessed(ctx); err != nil {
			return err
		}
	}