Lockdown Mode pallet.

[hbulgarini]

This PR implements a lockdown mode pallet inspired in the Moonbeam maintenance pallet implementation.

Specs can be found here: #144 (comment)

Closes #144

[hbulgarini]

The PR is ready to be reviewed now. Still needs to be confirmed, but it seems the chess pallet is the responsable for the wasm exceed during execution.

[kalaninja]

what is the reason for having it activated at the start?

[hbulgarini]

so we can have the chance to check that everything is fine before making the operations available.

[kalaninja]

I got it, but it doesn't seem to be responsibility of the pallet. This decision (on or off at the start) should be taken on a composition root level. My suggestion here is to add activated into the genesis config

	#[pallet::genesis_config]
	/// Genesis config for lockdown mode pallet
	pub struct GenesisConfig<T: Config> {
		pub activated: bool,
		pub _phantom: PhantomData<T>,
	}

	impl<T: Config> Default for GenesisConfig<T> {
		fn default() -> Self {
			Self { activated: ACTIVATED, _phantom: PhantomData }
		}
	}

	#[pallet::genesis_build]
	impl<T: Config> GenesisBuild<T> for GenesisConfig<T> {
		fn build(&self) {
			LockdownModeStatus::<T>::put(&self.activated);
		}
	}

this will also let you to set initial state for you tests

pub fn new_test_ext(activated: bool) -> sp_io::TestExternalities {
	let mut storage = system::GenesisConfig::default().build_storage::<Test>().unwrap();
	pallet_lockdown_mode::GenesisConfig::<Test> { activated, _phantom: PhantomData }
		.assimilate_storage(&mut storage)
		.unwrap();

	storage.into()
}

So that you don't need to activate lockdown mode first to deactivate it in the next step.

You can also add a test to make sure that the storage is initialized properly

#[test]
fn initialized() {
	[true, false].into_iter().for_each(|expected| {
		new_test_ext(expected).execute_with(|| {
			let lockdown_mode = LockdownModeStatus::<Test>::get();
			assert_eq!(lockdown_mode, expected);
		});
	});
}

[stiiifff]

@hbulgarini I think @kalaninja makes sense. If you can apply those changes, we should be good then.

[hbulgarini]

#172

[stiiifff]

@hbulgarini I think calling this pallet lockdown-mode would better represent the intent (I just found out that Apple has such a mode in iOS).

[hbulgarini]

@stiiifff @kalaninja @valentinfernandez1 , i have addressed all the comments.

[hbulgarini]

@kalaninja , i have to solve some conflicts on this branch because of the merge of the PR #165, could you please make sure that something was not incorrectly solved during the sync? Then if everything looks fine, you can just approve and merge. @stiiifff already approved it but due to the re-sync the approval was missed.

[stiiifff]

@hbulgarini approved but I think it'd be nice to implement @kalaninja 's suggestion (see above).

[stiiifff]

LGTM. Let's create a new issue with @kalaninja's improvement suggestion.