cli-server-restart-version-mismatch error

Nobody has claimed this yet.

### Search first

- [x] I searched and no similar issues were found

### What Happened?

Since the last nightly build, i get this issue in terminal every time a graph is loaded:

```
06:07:53.327 › {:cli-server-restart-version-mismatch {:port 36313, :owner-source :cli, :expected-revision "87f9aef-dirty", :current-revision "b6fec73-dirty", :host "127.0.0.1", :pid 490313, :root-dir "/home/davidak/logseq", :status :ready, :repo "logseq_db_Test"}, :line 347} null
```

### Reproduce the Bug

0. Start Logseq (you see it for the default loaded graph)
1. Switch Graph (you see it for every graph)

### Expected Behavior

No Errors!

### Screenshots

_No response_

### Files

_No response_

### Browser, Desktop or Mobile Platform Information

NixOS 25.11.10684.8fd9daa3db09
Logseq-linux-x86_64-2.0.1-alpha+nightly.20260624

### Additional Context

This is a regression. It did not happen in Logseq-linux-x86_64-2.0.1-alpha+nightly.20260623.

Please make sure this issue does not happen again.

### Are you willing to submit a PR? If you know how to fix the bug.

- [ ] I'm willing to submit a PR (Thank you!)

Start by reproducing the error on NixOS with nightly.20260624: launch Logseq with the default graph, then switch graphs and compare the behavior with nightly.20260623. Use the logged expected-revision, current-revision, port, and graph details to trace the CLI server restart version mismatch; done means no error appears during startup or graph switching.

Written by the indexing model from the issue text.