Raise a custom exception for partial dependence errors

[freddyaboulton]

Pull Request Description

Fixes #2539

Design document here .

---

After creating the pull request: in order to pass the release_notes_updated check you will need to update the "Future Release" section of docs/source/release_notes.rst to include this pull request by adding :pr:123.

[codecov]

Codecov Report

Merging #2604 (53b932b) into main (3a00d47) will increase coverage by 0.1%.
The diff coverage is 100.0%.

@@           Coverage Diff           @@
##            main   #2604     +/-   ##
=======================================
+ Coverage   99.9%   99.9%   +0.1%     
=======================================
  Files        295     295             
  Lines      26848   26894     +46     
=======================================
+ Hits       26802   26848     +46     
  Misses        46      46             

Impacted Files	Coverage Δ
evalml/exceptions/__init__.py	100.0% <ø> (ø)
evalml/exceptions/exceptions.py	100.0% <100.0%> (ø)
evalml/model_understanding/graphs.py	100.0% <100.0%> (ø)
...del_understanding_tests/test_partial_dependence.py	100.0% <100.0%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 3a00d47...53b932b. Read the comment docs.

[freddyaboulton]

Adding this try unfortunately makes the diff seem bigger than it is. I think it's important to catch all possible errors that this can raise so that users don't have to resort to except Exception or except PartialDependenceError, ValueError and hope that'll catch everything.

[freddyaboulton]

We have an outstanding issue to clean up partial dependence (#2502) so maybe this will be cleaner once we redesign this with this requirement in mind.

[chukarsten]

Very awesome, great work. Thanks for consolidating all these exceptions and making it super tidy. I think this will also help downstream consumers of EvalML a LOT.

[bchen1116]

LGTM! Thanks for cleaning this up and making it easier to tell users what the issue is. The error docs don't appear on the docs though, unless you explicitly type it into the search?

[bchen1116]

Ah, so this error isn't necessarily due to Partial Dependence, so it doesn't necessarily fall into the PartialDependenceError category? Got it

[freddyaboulton]

Yea it's caused by something else in sklearn. Raising here makes sure the final except catches it and then we can wrap as PartialDependenceError with the "all other errors code"

[freddyaboulton]

Also, about the docs page. Yea you have to click on the drop down on the left hand side and then click on exceptions.

The "API Reference" is a manually curated list of functions/classes but it's not exhaustive.

[ParthivNaresh]

Solid work, love the Enum implementation

[angela97lin]

I love this change. LGTM!

[angela97lin]

Can we combine this with the following check for "URL", "EmailAddress", "NaturalLanguage"? They all have the INVALID_FEATURE_TYPE code :)