[Bug] Subatomic - If PIRATE needs blockchain rescan, swaps will have issues completing

[satindergrewal]

Executed subatomic swap from Windows machine as Alice which did a swap with Bob node on Linux.
Alice buying PIRATE, selling KMD.
Bob buying KMD, selling PIRATE.

Alice got following in subatomic swaps log:

rel.KMD/KMD  openrequest 1480128384 status.0 (RBthCSgNLE3rwvAKce8JNSo7xDpxQEiRTX/zs1zqks0tergf6nk69evm6awte4xmhf2fd9epnv946vzlhfxztkls6a9lyfmuafda00krvkvj0xagp)
1480128384 openrequest.3045915584 -> (0133f63a3d4ae8db7e9efe8b8702e10ecc9ef44901dd2321c92132889bc6656b4e)
1480128384 iambob.0 (PIRATE/KMD) channelapproved origid.1480128384 status.1
1480128384 approvalid.3951842080 (0133f63a3d4ae8db7e9efe8b8702e10ecc9ef44901dd2321c92132889bc6656b4e)
1480128384 iambob.0 (PIRATE/KMD) incomingchannel status.2
1480128384 got txid.a1fc91cae5963a0cbb7955b149ea0924195bb6999e0af1ce2fdde34d25cfe8b9
1480128384: 0.01000000 KMD -> RQPZrM4yQaTZpEuoGmcGwzE4SaG2Tn9QiB, paymentid[0] 3706397792
1480128384 iambob.0 (PIRATE/KMD) incomingpayment status.4
1480128384 alice waits for PIRATE.0000000000000000000000000000000000000000000000000000000000000000 to be in mempool (0.05000000 -> zs1zqks0tergf6nk69evm6awte4xmhf2fd9epnv946vzlhfxztkls6a9lyfmuafda00krvkvj0xagp)
1480128384 SWAP INCOMPLETE, waiting on PIRATE.0000000000000000000000000000000000000000000000000000000000000000
1480128384 closedid.2261577536
subatomic_channel_alice (KMD/KMD) 1480128384 1480128384 with 0.01000000 1000000
start subatomic_loop iambob.0 PIRATE -> KMD, 1480128384 1000000 3045915584
sendtoaddress RQPZrM4yQaTZpEuoGmcGwzE4SaG2Tn9QiB 0.01000000 txid.(a1fc91cae5963a0cbb7955b149ea0924195bb6999e0af1ce2fdde34d25cfe8b9)
dpow_broadcast.(incomplete/a1fc91cae5963a0cbb7955b149ea0924195bb6999e0af1ce2fdde34d25cfe8b9) [ ] a1fc91cae5963a0cbb7955b149ea0924195bb6999e0af1ce2fdde34d25cfe8b9 error.(-1)

For PIRATE, the txid received is 0000000000000000000000000000000000000000000000000000000000000000 at Alice end.

On Bob's end following are the logs:

Bob's PIRATE debug.log shows:

2020-07-04 16:03:19 ERROR: AcceptToMemoryPool: joinsplit requirements not met
2020-07-04 16:03:19 opid-ce62364f-c6e9-4baf-8583-fa6ca5920dfe: z_sendmany finished (status=failed, error=18: bad-txns-joinsplit-requirements-not-met)

subatomic should have the condition to handle a failed z_sendmany output.