PR: Support all real number dtypes in dataframe editor

[jitseniesen]

Description of Changes

• Wrote at least one-line docstrings (for any new functions)
• Added unit test(s) covering the changes (if testable)

• Included a screenshot or animation (if affecting the UI, see Licecap)

This is a small PR which uses the new function pandas.api.types.is_any_real_numeric_dtype() to decide whether a Pandas dtype is a real number and thus gets the background colouring of numbers in the Dataframe editor. This allows us to supports the nullable integer types and also numeric types that were previously not supported like numpy.int16. A test is also included.

This PR imports pandas inside a function. As far as I can see, no part of Spyder explicitly imports pandas but I think that pandas is always imported when a Dataframe editor is opened, so this should not be a problem.

Screenshot after:

Screenshot before:

These are produced using the following code, from issue #21022:

import pandas as pd

vals = [1, 2, 3, 4, 5]
vals_na = [1, 2, 3, None, 5]

df = pd.DataFrame({
    'old': pd.Series(vals),
    'old_na': pd.Series(vals_na),
    'new': pd.Series(vals, dtype='Int64'),
    'new_na': pd.Series(vals_na, dtype='Int64')
})

Issue(s) Resolved

Fixes #21222

Affirmation

By submitting this Pull Request or typing my (user)name below,
I affirm the Developer Certificate of Origin
with respect to all commits and content included in this PR,
and understand I am releasing the same under Spyder's MIT (Expat) license.

I certify the above statement is true and correct: Jitse Niesen

[ccordoba12]

A couple of small suggestions for you, the rest looks good to me.

[jitseniesen]

A couple of small suggestions fo you, the rest looks good to me.

Thanks, I applied all the suggestions.

[ccordoba12]

Thanks @jitseniesen!