DOC: Add glossary to the documentation

[datapythonista]

We often use terms in the documentation that may not be clear for the users. For example:

• label (see Add informal definition of "label" in note #27005)
• list-like (see Clearly document what a list-like is #21784)

scikit-learn added a glossary to their documentation (https://scikit-learn.org/stable/glossary.html), and I think this would also be useful for pandas. Not sure the exact terminology to start with, but feel free to comment here with your ideas, and we can work on this once we have a first list.

[TomAugspurger]

My main concern here is duplication of information. Hopefully we could pair adding a glossary with a cleanup of existing definitions.

A few other terms that may be worth having entries for.

• alignment
• dtype
• missing value / missing data (we already have a section for this)
• extension array
• method chaining
• offset (time series)

[WillAyd]

Throwing a few more into the ring for consideration:

• Datetime-like
• Arraylike
• Path or Buffer
• Numeric (or at least mention this doesn't include Decimal)