[R4R] #499 failed blocking should not be regarded as closed order

[cryptom-dev]

Description

failed blocking should not be regarded as closed order

Rationale

#499

Example

N/A

Changes

1. don't delete order in dexkeeper.orderInfoForPub for failedblocking
2. change payload of failedblocking order to failed message (not rely on orderInfoForPub anymore)
3. cleaner and safer code to maintain closed and open order. The closed concept is consistent with QS
4. fix new added order status field is not correctly involved in orderbook calculation

Preflight checks

• build passed (make build)
• tests passed (make test)
• integration tests passed (make integration_test)
• manual transaction test passed (cli invoke)

Already reviewed by

...

Related issues

#499

[darren-liu]

return msg.Status == orderPkg.IocNoFill || (msg.CumQty == 0 && msg.Status == orderPkg.Expired)

[cryptom-dev]

return msg.Status == orderPkg.IocNoFill || (msg.CumQty == 0 && msg.Status == orderPkg.Expired)

We cannot change to that.
Partial filled IOC order would also publish a IocNoFill status to indicates it's closed. And this method is statistic whether this ioc no fill should be charged.

More background: we publish:
for no fill ioc: Ack, IocNoFill
for partial or fully fill ioc: Ack, PartialFill (with a counterpart1)...PartialFill (with counterpart2), IocNoFill