Feature: Add load_CB_speeches function

[robin-mader-bis]

This Pull Request adds a load_CB_speeches function for downloading the soon to be published Central Bankers Speeches dataset from www.bis.org.

Summary of changes:

• Add load_CB_speeches function and related documentation
• Add internals module, intended for functions which are intended for package internal use only
• Add settings module, for package internal setting constants
• Add citation for Central Bankers speeches dataset

[dkgaraujo]

Many thanks @robin-mader-bis. The only thing I'm missing is the new rendered datasets.html. It's a matter of simply rendering the corresponding qmd file: quarto render 00_datasets.qmd. Please include also the resulting docs\datasets.html file in the PR. The rest shoul;d be good to go, pending also the comments above.

[robin-mader-bis]

@dkgaraujo My bad, I thought this would be added automatically by the CI. Should I also include other files that are changed during the render process, or only the docs\datasets.html?

[dkgaraujo]

@dkgaraujo My bad, I thought this would be added automatically by the CI. Should I also include other files that are changed during the render process, or only the docs\datasets.html?

No problem. The CI part will be added later, but for now the PRs need to render all the files automatically.

Since the only "numbered" quarto file you changed was 00_datasets.qmd, there is no need to re-render the whole project, just that file.

[dkgaraujo]

One more thing I uncovered testing on python 3.9: we need to add the folling import at the very beginning of datasets.py, similar to estimators.py:

from __future__ import annotations  # Allows forward annotations in Python < 3.10

This allows for arguments that have multiple types.

[robin-mader-bis]

@dkgaraujo Good catch, I added the import as suggested