fix the Unauthorized error of full node caused by mempool tx

[AaronZgl]

Description

closes: #8510

1. If a tx passes anteHandler of a full node but fail to pass anteHandler of all validators cause the minGasPrice of the full node is less than all validators, the tx will not be included in any block and only exist in the memory pool of the full node and be rechecked after per block.
2. At this point, if the user sends another tx to the full node, the new tx will get the unauthorized error because the correct sequence of user will be added in NewIncrementSequenceDecorator when the old tx was rechecked. So,the NewIncrementSequenceDecorator function should be skipped in recheck.
3. If both NewIncrementSequenceDecorator and NewSigVerificationDecorator were skipped in recheck, the old tx would always exist in the memory pool. And if NewSigVerificationDecorator is not be skipped in recheck, the old tx would fail to verify signature and be removed when the sequence of user was updated.

So the NewIncrementSequenceDecorator function should be skipped in recheck but not NewSigVerificationDecorator.

---

Before we can merge this PR, please make sure that all the following items have been
checked off. If any of the checklist items are not applicable, please leave them but
write a little note why.

• Targeted PR against correct branch (see CONTRIBUTING.md)
• Linked to Github issue with discussion and accepted design OR link to spec that describes this work.
• Code follows the module structure standards.
• Wrote unit and integration tests
• Updated relevant documentation (docs/) or specification (x/<module>/spec/)
• Added relevant godoc comments.
• Added a relevant changelog entry to the Unreleased section in CHANGELOG.md
• Re-reviewed Files changed in the Github PR explorer
• Review Codecov Report in the comment section below once CI passes

[github-actions]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[alexanderbez]

This is not correct. We can and should bypass signature verification on rechecking.

[aaronc]

Blocking until we get consensus on the correct behavior. @alexanderbez ?

[alexanderbez]

The NewIncrementSequenceDecorator change is correct. In fact, the fact that it's not there is a regression. See the original PR: https://github.com/cosmos/cosmos-sdk/pull/5950/files.

The NewSigVerificationDecorator I think is NOT correct. One of the entire points of ReCheckTx is the fact that we can skip sig verification.

[aaronc]

The NewIncrementSequenceDecorator change is correct. In fact, the fact that it's not there is a regression. See the original PR: https://github.com/cosmos/cosmos-sdk/pull/5950/files.

It was reverted in #6291. Do you have any context for that @alexanderbez ?

[github-actions]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[alexanderbez]

It was to allow multiple txs in the same block from a sender which has always been a point of confusion for me.

[tac0turtle]

pinging for an update here

[github-actions]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.