Add VTGate drain flag

[frouioui]

Description

This PR adds a new flag to VTGate to allow for all the MySQL connections to be drained when terminating VTGate. The drain wait until all the idle connections have disconnected (or the timeout expires, whichever happens first). When the server goes into OnTerm state, new connections will stop being accepted, and the load balancer will connect new application sessions to other vtgates. Once all the idle app connections have drained (because of the max lifetime of 1 hour of a session on the client side), the vtgate will exit.

Checklist

• "Backport to:" labels have been added if this change should be back-ported to release branches
• If this change is to be back-ported to previous releases, a justification is included in the PR description
• Tests were added or are not required
• Did the new or modified tests pass consistently locally and on CI?
• Documentation was added or is not required

Deployment Notes

[vitess-bot]

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

• Ensure that the Pull Request has a descriptive title.
• Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

• Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

• Apply the release notes (needs details) label if users need to know about this change.
• New features should be documented.
• There should be some code comments as to why things are implemented the way they are.
• There should be a comment at the top of each new or modified test to explain what the test does.

New flags

• Is this flag really necessary?
• Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

• Each item in Jobs should be named in order to mark it as required.
• If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

• Protobuf changes should be wire-compatible.
• Changes to _vt tables and RPCs need to be backward compatible.
• RPC changes should be compatible with vitess-operator
• If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
• vtctl command output order should be stable and awk-able.

[harshit-gangal]

With Application side connection pooling this will most likely not change and VTGate can remain in the same state before being shut down.

[arthurschreiber]

I'm trying to understand how this differs from / interacts with the connection shutdown functionality we added some while ago?

When Listener.Shutdown is called (just a few lines above what you added), vtgate starts explicitly closing open connections when a QUERY or PING MySQL command is received, and the connection is not currently inside of a transaction. The error code we send back is 1053 (Server shutdown in progress), which signals to clients that they need to re-connect.

For connections that are inside of a transactions, we allow them to continue on until the end of the grace period is reached (at which point we explicitly terminate their connection). Same for idle connections - they get terminated explicitly if the grace period is reached.

So what is this PR attempting to fix / change exactly? 🤔

[arthurschreiber]

Also, I think the existing logic to wait for connections to become idle is not working correctly / doesn't make sense.