Sovereign Sequencer v1: decide on how to deal with Escape Hatch

[ImJeremyHe]

I think we can add a flag indicating that we are running the sovereign sequencer and leave the escape hatch as it is. This might increase some mental burden, but right now, I believe it is necessary until we thoroughly understand the usage of our integration.

We can tamper with the L2 message type if we are running the sovereign sequencer so that all components can handle the L2 messages more conveniently.

[sveitser]

@ImJeremyHe can you explain to me why we need that flag?

[ImJeremyHe]

After working a bit on the sovereign things I think I can come up with a more detailed explanation.

Right now, our nitro integration are handling 3 kinds of messages:

• Espresso Messages. These messages are fetched from hotshot directly when HotShot is up. These messages need to be validated by the hotshot validation pipeline, like the namespace and the hotshot block continuity.
• Centralized Messages. These messages are fetched from the centralized sequencer when HotShot is down or we are not running with any espresso mode.
• Sovereign Messages. These messages are created originally from the centralized sequencer, but later will be sent to the HotShot. And these messages should be validated as the espresso messages except for the hotshot block continuity.

So you can see the sovereign messages actually have broken some of our previous conventions. So having a specific flag for the sovereign should be reasonable, or it is hard for us to maintain everything. Of course, if we finally set up a determined usage of how the integration rollups should act, we can reduce these flags.