Bridges testing improvements

[bkontur]

This is a global summary issue for all testing-related matters concerning Bridges.
(all issues will be re-visited and adjusted)

Runtime testing

• Try to replace our custom integrity tests (explicitly callable) with similar Substrate tests parity-bridges-common#2612
• Add test for MMR::on_initialize() vs Session::on_initialize() parity-bridges-common#2579
• Better testing for wrapped runtimes parity-bridges-common#2453
• dry_run_xcm support for BridgeHubs #5708
• Bridge Permissionless Lanes - more tests #5552

Live testing (for testnets)

• Revamp and improve testing for Rococo <> Westend bridge parity-bridges-common#2640
• TODO: prepare tests for permissionless lanes
• TODO: explore zombienet-bite

Fellows testing

• Add Bridges integration tests to the Fellowship CI polkadot-fellows/runtimes#544
• Bridges polkadot-sdk/fellows zombienet tests nits parity-bridges-common#3016
• Integration tests and setup multiple relayers for "free header" feature for polkadot-fellows parity-bridges-common#3017
• TODO: prepare tests for permissionless lanes

Integration-tests - xcm-emulator related

• Bridge xcm emulator testing improvements parity-bridges-common#2849
• ? Equivocations detection loop: add integration tests parity-bridges-common#2610

Chopsticks support

• Bridges: enable chopsticks to emulate passing XCM messages over bridge #4793
• Test P<>K bridge-init calls using chopsticks #3522
• Mock grandpa and beefy rpc to allow P<>K bridge-init calls using chopsticks AcalaNetwork/chopsticks#681
• feat: add grandpa justifications mock rpc AcalaNetwork/chopsticks#695
• Add support for P<>K bridge AcalaNetwork/chopsticks#741
• Add chopsticks tests for XCM use cases parity-bridges-common#2953

Others

• How to enable easier integration/e2e testing for parachains which want to use P<>K bridge?
• Build bridges testing framework #3242
• Run Bridges Zombienet tests (Polkadot SDK, Polkadot Fellows) for Relayers PRs (bridges-common CI) parity-bridges-common#3110
• Investigate and rewrite bridges zombienet tests with zombienet-sdk #7665

[bkontur]

@serban300 @franciscoaguirre I think the dry-run and Chopsticks way has the same common problem, because there could possibly be DescendOrigin added here when dispatching on bridged chain. So, if we need to add some runtime api or whatever for that, it should cover both cases.