Resuming indexing consistently fails with "Could not find outpoint in index"

[veryordinally]

After we implemented #496, resuming indexing consistently fails on the main chain starting from block heights above 150000 or so, that is if we run ord index --height-limit 150000 (starting with no index) and then run ord index again, it immediately fails with:

[2022-09-09T16:31:55Z INFO  ord::index] Block 149999 at 2011-10-20 13:40:14 UTC with 17 transactions…
[2022-09-09T16:31:55Z INFO  ord::index] Block receiver disconnected: sending on a closed channel
error: Could not find outpoint in index

Resuming indexing works for smaller block heights.

It seems the index is inconsistent in some way. I had not seen this behavior before we implemented #496

[casey]

Fixed! I made an off-by-one error in #495.