PERF: Improve performance when accessing GroupBy.groups

[phofl]

• closes #xxxx (Replace xxxx with the GitHub issue number)
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

df = DataFrame(np.random.randint(1, 100, (5_000_000, 5)))
df.groupby([0, 1]).groups

# main
%timeit df.groupby([0, 1]).groups
1.68 s ± 7.52 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
# pr
%timeit df.groupby([0, 1]).groups
504 ms ± 9.24 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

[jbrockmendel]

can we use a specific isinstance check instead?

[phofl]

Ah yes, we can go the other way round

[mroeschke]

Thanks @phofl