adding network removed event

[haider-rs]

Description

closes: #1156

Type of change

• New feature (non-breaking change which adds functionality)

Tests

• test_add_and_remove_network

Code review prechecks:

• Code follows the style guidelines of this project
• Code has been self-reviewed
• Inline comments have been added for each method
• I have made corresponding changes to the documentation
• Code changes introduces no new problems or warnings
• Test cases have been added

[penumbra23]

Why is the analog-gmp submodule updated in almost every PR?

[haider-rs]

@penumbra23 yes thats because testnet branch uses a different version of analog-gmp and main uses a different so everytime we open a PR the analog-gmp branch is old and we forget to revert back to that commit.

[FlorianFranzen]

So what happens with all the other data of that network on chain?

Would that mean if I remove a network and add one back with the same chain id, I would still have all the previous tasks and shards assigned?

[dvc94ch]

It doesn't really matter since it's going in the Testnet branch. Adeel requested it, so was either discussing why it doesn't make sense for 2h or do it in 10min

[FlorianFranzen]

@dvc94ch It does matter very much, because it is going to testnet. Do we for example have tests what happens when tasks are schedule on non-existing networks? Maybe just running try-runtime in follow mode would be good.

Did anybody do any analysis, where network ids are used or iterated over?

[dvc94ch]

Bran told Adeel it was nonsensical. I wasn't in the meeting. Please discuss the assignment with Adeel...

[haider-rs]

Agree with Florian, If the PR made any unexpected storage corruption there is no going back except from complete reset.
cc: @0x1100010010