Add SPLADEv3 into Pyserini

[lintool]

Let's incorporate SPLADEv3 into Pyserini.

Start here: https://dl.acm.org/doi/10.1145/3626772.3657862

This paper has SPLADE++ ED (the previous version).

It'd be great to have someone repro SPLADE v3 from this repo: https://github.com/naver/splade

A start would be to augment the existing onboarding docs. I.e., here, we build BGE-base for NFCorpus: https://github.com/castorini/pyserini/blob/master/docs/experiments-nfcorpus.md

Let's figure out how to do the equivalent for SPLADEv3?

From there, we can build out SPLADEv3 bindings for all of BEIR, MS MARCO, etc.

[clides]

I'll work on this

[lintool]

Thanks @clides - initial onboarding doc merged in this PR castorini/pyserini#2085

[clides]

the current workflow goes:

1. write a script that uses SpladeQueryEncoder (a module from pyserini.encode) to encode the corpus
2. use pyserini to index the embeddings
3. perform retrieval on the corpus (I think the queries are encoded in this step, however, ill need to take a deeper look at the code later)
4. eval the model