refactor: replace ValidationError string with PayloadValidationError enum

[TechieBoy]

closes #2720

Do we also need to create seperate PayloadValidationError variants for every consensus/Execution error?

https://github.com/paradigmxyz/reth/blob/5039122c356f20d36e3d5979f0da97a2fc40ac75/crates/consensus/beacon/src/engine/mod.rs#L743-L746

[codecov-commenter]

Codecov Report

Merging #2803 (a29c38d) into main (91dd782) will decrease coverage by 27.12%. The diff coverage is 0.00%.

Impacted Files	Coverage Δ
crates/rpc/rpc-types/src/eth/engine/payload.rs	30.89% <0.00%> (-56.50%)	:arrow_down:

... and 257 files with indirect coverage changes

Flag	Coverage Δ
integration-tests	15.88% <0.00%> (-0.01%)	:arrow_down:
unit-tests	33.98% <0.00%> (-30.12%)	:arrow_down:

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
reth binary	10.74% <ø> (-15.61%)	:arrow_down:
blockchain tree	69.93% <ø> (-11.33%)	:arrow_down:
pipeline	49.75% <ø> (-36.93%)	:arrow_down:
storage (db)	54.01% <ø> (-19.48%)	:arrow_down:
trie	58.94% <ø> (-35.73%)	:arrow_down:
txpool	30.28% <ø> (-18.71%)	:arrow_down:
networking	45.03% <ø> (-32.80%)	:arrow_down:
rpc	33.81% <0.00%> (-24.31%)	:arrow_down:
consensus	36.74% <ø> (-26.62%)	:arrow_down:
revm	18.87% <ø> (-15.99%)	:arrow_down:
payload builder	6.83% <ø> (ø)
primitives	49.89% <ø> (-38.39%)	:arrow_down:

[mattsse]

okay, so let's get rid of all the deserialize impls?

[Rjected]

okay, so let's get rid of all the deserialize impls?

yeah

[onbjerg]

@Rjected @mattsse what's the path forward here?

[Rjected]

After thinking about it, to remove the Deserialize impl we would have to remove it from a bunch of other engine API types. This might be fine, since we only actually serialize those types afaik. Including a more rich type in an error field that needs to be deserialized requires parsing the error, which is why the Deserialize impls are so involved.

I think it's fine to remove the Deserialize impls, but it's a bit more work than I thought so that would be the next steps here.

[mattsse]

I thought about this a bit,

and I no longer support replacing string with an enum, because this just further complicates things, it's okay to treat this as a simple error message.

I kept the tests though, ty!