feat: adapt eth_sender tests for running both in Validium mode and Rollup mode

[toni-calvin]

Description

Closes #118

This Pull Request addresses the need to support and test a new type of SenderTester, specifically the "validium" tester, in addition to the default "rollup" tester. In a previous PR, we introduced different testers, but the default was set to "rollup" without adequate testing for the "validium" tester.

The main objective of this PR is to enhance testing coverage for the "validium" tester. The existing test structure has been adjusted to accept a SenderTest as a parameter, allowing for greater flexibility in testing various tester types. Additionally, new methods have been added to validate the tests specifically for each tester type, ensuring thorough and targeted testing for both "rollup" and "validium" testers.

Changes Made

• Adjustment of the l1_batch_commit_generator to accommodate the unique requirements of the "validium" tester, as it is the only differing aspect between tester types.
• Adaptation of the existing test structure to accept a SenderTest as a parameter.
• Addition of methods that instantiate one of each type of SenderTester (including "rollup" and "validium") and call the existing tests, now with the Tester as a parameter.

How to Test

• For testing, inside the root folder of the project run:

zk test rust --package zksync_core --lib -j 1 -- eth_sender::tests

[ilitteri]

Can't we just do this? (We'd get rid of the getter method)

Suggested change

	let l1_batch_commit_data_generator: Arc<dyn L1BatchCommitDataGenerator> =
	tester.aggregator.get_l1_batch_commit_data_generator();
	let l1_batch_commit_data_generator: Arc<dyn L1BatchCommitDataGenerator> =
	tester.aggregator.l1_batch_commit_data_generator.clone();

[toni-calvin]

In order to do that, we should expose public the aggregator field inside the EthTxAggregator and the field l1_batch_commit_data_generator.

[ilitteri]

The base branch should be feat_validium_pubdata_abstraction.

[ilitteri]

The PR is looking good, great work! Here I leave some comments:

• Let's add a description to the PR, given that there's an issue associated we can mention Resolves #X or Closes #X. Then if some additional information is needed by the reviewer it'd be good to add it in subsequent sections (like how to run the tests you're changing).
• Let's move the auxiliary functions (the abstractions) to a test_helpers module. If I'm not wrong, by doing so we an get rid of the prefix _ and call the methods like test_helpers::xxx after importing the module test_helpers.
• Regarding the get_l1_batch_commit_data_generator method: I see why you added it, but I think we can avoid needing it (in terms of design it'd be ok to have it as a method for Aggregator by maybe not for EthTxAggregator, we'll still remove it from both though). As it is only meant for testing purposes we can make the method public for the crate or super if necessary, even better, we can make the struct fields public for the crate or super; this won't be the case either because it'd be more difficult to be accepted in the main PR.
  I saw in the tests that it is only consumed by commit_l1_batch to instantiate an AggregatedOperation::Commit which instantiates a CommitBatches struct that will be the final consumer of l1_batch_commit_data_generator. This is great news because this means that the EthTxAggregator's Aggregator that is in the tester is not the one in charge of creating an AggregatedOperation::Commit for later committing, therefore we can pass l1_batch_commit_data_generator as a parameter to the functions along with the tester and get rid of using the method and roll it back.

[ilitteri]

LGTM! Great work!