feat: Port evaluation harness from haystack-experimental

[shadeMe]

Proposed Changes:

Port the evaluation harness from the experimental package.

How did you test it?

Unit tests

Notes for the reviewer

The source feature branch doesn't originate from a fork, so you can directly push to branch if need be.

We should ensure that we have the usage docs for the harness and its related classes ready so that both the code and the docs are merged at the same time, i.e., during the same release.

@dfokina Feel free to review the docstrings as well.

Checklist

• I have read the contributors guidelines and the code of conduct
• I have updated the related issue with new insights and changes
• I added unit tests and updated the docstrings
• I've used one of the conventional commit types for my PR title: fix:, feat:, build:, chore:, ci:, docs:, style:, refactor:, perf:, test:.
• I documented my code
• I ran pre-commit hooks and fixed any issue

[coveralls]

Pull Request Test Coverage Report for Build 12200323610

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• 33 unchanged lines in 4 files lost coverage.
• Overall coverage increased (+0.4%) to 90.707%

---

Files with Coverage Reduction	New Missed Lines	%
components/retrievers/sentence_window_retriever.py	1	98.08%
components/preprocessors/nltk_document_splitter.py	3	97.35%
components/converters/pypdf.py	6	93.55%
components/converters/azure.py	23	89.38%

Totals
Change from base Build 12086061543:	0.4%
Covered Lines:	8453
Relevant Lines:	9319

---

💛 - Coveralls

[bilgeyucel]

Can we add the DocumentNDCGEvaluator to the supported metrics as we port it?
Another small feature request from the community: deepset-ai/haystack-experimental#74 (reply in thread)

[davidsbatista]

• IMO the abstract class BaseEvaluationRunResult is unnecessary, and we could just have a classEvaluationRunResult.
• In the class EvaluationRunResult(BaseEvaluationRunResult) we should remove all the uses of DataFrame, i.e.: output .CSV instead of DataFrames and remove pandas as a dependency.

[davidsbatista]

adding @anakin87 since he's working on removing pandas/DataFrames dependencies

[anakin87]

@davidsbatista thanks for the ping.

In general, my idea would be to have pandas as an optional dependency like many others.
So, if it makes sense in this case, we can leave it and making it optional in the future, wrapped in a Lazy Import block.

[davidsbatista]

@anakin87, I don't oppose removing pandas from the EvalHarness and output .csv; it could make this code a bit simpler.

But if we want to keep it optional, some functionality like score_report() (and others) either only work with pandas installed, or we need to extend them to have the option to output either CSV or DataFrame.