feat: flush leader region before downgrading

[WenyXu]

I hereby agree to the terms of the GreptimeDB CLA.

Refer to a related PR or issue link (optional)

What's changed and what's your

This change helps the system handle region migration more smoothly. By flushing the leader region before downgrading it, we can:

• Minimize the time window where the region is not writable
• Reduce the impact on write operations during region migration

Changes

• Added a new FlushLeaderRegion state in the region migration procedure
• Modified the region migration flow to flush the leader region before downgrading it
• Added new metrics for tracking flush leader region operation duration
• Added new test cases for the flush leader region functionality

PR Checklist

Please convert it to a draft if some of the following conditions are not met.

• I have written the necessary rustdoc comments.
• I have added the necessary unit tests and integration tests.
• This PR requires documentation updates.
• API changes are backward compatible.
• Schema or data changes are backward compatible.

[fengjiachun]

LGTM

[MichaelScofield]

Why not make the leader flushing inside downgrading it? That way you can avoid addition writes between flush and downgrade.

[WenyXu]

Why not make the leader flushing inside downgrading it? That way you can avoid addition writes between flush and downgrade.

By flushing the leader region before downgrading it, we can minimize the time window where the region is not writable.