refactor(cli): refactor to esm

[dk1a]

Replaces #397

From my testing this is the best solution for table autogeneration, and should be neutral for current use-cases.

Latest main should be merged into v2 before this PR (network's createFaucetService is duplicated in both)

There're no significant logic changes here, but I remove some seemingly deprecated commands to do less refactoring (deploy, diamond-abi, sync-art).
Added reuseComponents to deploy-contracts since it was only in deploy.

I've started going from no bundler+ts-node-esm+moduleResolution nodenext (which doesn't work well yet) to eventually tsup with some inlining+node+moduleResolution node. The changes that make the 1st more extreme version work could be removed, but they are a strict improvement and could simplify things in the future.
Notably:

• .js suffixes for files
• services changes
• tsc --moduleResolution nodenext
• execLog refactor was just me testing stuff, but it should be a strict improvement worth keeping

@holic create-mud was working without @latticexyz/utils seemingly because it was cli's dependency, even though cli didn't (and couldn't) use it. And since it's std-client's peerDep it's best to add it anyways.

@authcall importSuffix=.js helps with future compatibility with es module resolution. (It'd be even better if ts-proto and protobufjs would replace long with bigint entirely protobufjs/protobuf.js#1557).

[alvrs]

@Kooshaba this is not used in sky strife anymore, is it?

[alvrs]

chatted on discord, seems like sky strife is actually still using this command, so we shouldn't remove it for now

[alvrs]

This looks good to me! Tested a couple commands locally and everything still works as expected. Good to go once main was merged into v2 to avoid a conflict with the changes in createFaucetService

[alvrs]

unfortunately we actually do need sync-art, see #417 (comment)

[alvrs]

(or maybe sync-art could be a standalone script outside the mud cli)

[holic]

(or maybe sync-art could be a standalone script outside the mud cli)

Yeah this would be my preference, since sync-art feels very Phaser specific (if not Sky Strife specific)

[dk1a]

@alvrs rebased; test some commands again just to make sure it still works.
There were quite a lot of conflicts, some rather indirect. I think to avoid dealing with those often it'd be best to avoid cli changes in main. Or merge v2 into main, if we're ok with a few breaking cli changes and a bunch of unfinished packages there.

(or maybe sync-art could be a standalone script outside the mud cli)

Yeah this would be my preference, since sync-art feels very Phaser specific (if not Sky Strife specific)

It makes sense to put that script into sky strife repo. It's small, self-contained and very project-specific. Leaving sync-art removed for now, unless you disagree

[alvrs]

There were quite a lot of conflicts, some rather indirect. I think to avoid dealing with those often it'd be best to avoid cli changes in main. Or merge v2 into main, if we're ok with a few breaking cli changes and a bunch of unfinished packages there.

Yes, I think that makes sense, I'm all for merging v2 into main as soon as possible.

It makes sense to put that script into sky strife repo. It's small, self-contained and very project-specific. Leaving sync-art removed for now, unless you disagree

Agreed

[alvrs]

The gas-report cli is failing because the gasreport script in store still refers to cli/dist/index.js instead of cli/dist/mud.js - we could probably change that now to use the released @latticexyz/cli gas-report command

[dk1a]

The gas-report cli is failing because the gasreport script in store still refers to cli/dist/index.js instead of cli/dist/mud.js - we could probably change that now to use the released @latticexyz/cli gas-report command

Right, missed that one