feat: clean cache

[evilrobot-01]

Adds a simple command which allows a user to clean the cache by selecting those cached artifacts which are no longer required.

Note: uses 'clean' so that we can potentially add a default implementation of 'pop clean' to wrap 'cargo clean'. No real benefit apart from fewer characters to type.

Closes #78

[codecov]

Codecov Report

Attention: Patch coverage is 74.59016% with 93 lines in your changes missing coverage. Please review.

Project coverage is 66.00%. Comparing base (03aecfe) to head (2973f9f).

@@            Coverage Diff             @@
##             main     #216      +/-   ##
==========================================
+ Coverage   65.37%   66.00%   +0.63%     
==========================================
  Files          38       40       +2     
  Lines        5048     5413     +365     
  Branches     5048     5413     +365     
==========================================
+ Hits         3300     3573     +273     
- Misses       1229     1277      +48     
- Partials      519      563      +44     

Files	Coverage Δ
crates/pop-cli/src/main.rs	45.33% <ø> (ø)
crates/pop-cli/src/style.rs	82.14% <ø> (ø)
crates/pop-cli/src/commands/mod.rs	0.00% <0.00%> (ø)
crates/pop-cli/src/commands/clean.rs	81.72% <81.72%> (ø)
crates/pop-cli/src/cli.rs	69.94% <69.94%> (ø)

[evilrobot-01]

Need #217 merged so that I can complete the implementation of mocking for testing the implementation of the command.

[AlexD10S]

Remove #[cfg(feature = "parachain")] and support removing folders.

A nice-to-have feature would be the option to clean all with an all option in the multiselect or a --all flag. However, this feature is not essential for this PR. If it is not included, I will create a new issue with the good first issue tag for new contributors.

[evilrobot-01]

Note: have wrapped the cliclack usage in a trait in an effort to enable testing of interactions and improve coverage in the pop-cli crate. Its a manual mock so isnt perfect, but does enable some testing of command logic. Note also that the cli should not be defined on the command as it is not state applicable to the command, but is fine for an initial iteration. This should be refactored into a command handler in the future which injects the required dependencies for a cleaner implementation.

[evilrobot-01]

Remove #[cfg(feature = "parachain")] and support removing folders.

Why do we need to remove folders? To my knowledge no folders are created here.

A nice-to-have feature would be the option to clean all with an all option in the multiselect or a --all flag.

Would be nice to have but afraid I dont have the bandwidth to add this now.

[AlexD10S]

Looks good to be merged now.
I've created two new issues to address in separate PRs (#224 and #225).
Additionally, as per your comment, we'll need to refactor the cliclack wrapper and integrate it with the rest of the commands.

[evilrobot-01]

Additionally, as per your comment, we'll need to refactor the cliclack wrapper and integrate it with the rest of the commands.

I think we should address the cli parameter on the command as a separate PR first, as it is a dependency rather than command state, before moving all the commands to this type of implementation. I think it will be much easier that way, at least in terms of reviews.