refactor to rdFingerprintGenerator

[zhu0619]

Changelogs

This PR addresses the fingerprint deprecation warnings, see issue #225.

• Refactoring fingerprint atom pair, rdkit, morgan, topological, and their count fingerprints to FingerprintGenerator.
• ecfp-count now use the ECFP-6, to stay consistent.
• added new decorator function dm.no_rdkit_log which disable RDKit log for a whole function.
• set minimum rdkit version to avoid pandas based error for this version of datamol.

---

Checklist:

• [x] Was this PR discussed in an issue? It is recommended to first discuss a new feature into a GitHub issue before opening a PR.
• [x] Add tests to cover the fixed bug(s) or the new introduced feature(s) (if appropriate).
• [ ] Update the API documentation is a new function is added, or an existing one is deleted.
• [x] Write concise and explanatory changelogs below.
• [x] If possible, assign one of the following labels to the PR: feature, fix or test (or ask a maintainer to do it for you).

---

The rdFingerprintGenerator.GetCountFingerprint can be used to compute the count based fingerprint. The argument countSimulationindicates whether to compute the count fingerprint by bites simulation for computing efficiency. See more in this blog. I assume in datamol, we want to compute the exact count-based fingerprint. Therefore, countSimulation should be set to False.

[zhu0619]

Since it's still draft, I can wait.

The tests are failing, probably due to the different set arguments in the fingerprint generators. @maclandrol I need your help with revising default parameters. And I think the values in the tests for fingerprints need to be updated.

[maclandrol]

Since it's still draft, I can wait.

The tests are failing, probably due to the different set arguments in the fingerprint generators. @maclandrol I need your help with revising default parameters. And I think the values in the tests for fingerprints need to be updated.

Ok, will do.

[maclandrol]

@zhu0619, the default parameters were wrong. They are fixed now.

Still have to address this: https://github.com/danielfrg/mkdocs-jupyter/issues/200

Ideally in the next release, we should also have a minimum version for rdkit. As the pandas error still exists.

[maclandrol]

I also replaced seaborn by pure matplotllib in one of the notebook. Was getting an error and not sure how.

Feel free to merge anytime.