PERF: Improve performance of groupby.size

[rhshadrach]

• closes PERF: groupby.value_counts #51750

size = 10_000_000
df = pd.DataFrame(
    {
        "a": np.random.randint(0, 10000, size),
        "b": np.random.randint(0, 100000, size),
        "c": np.random.randint(0, 1000000, size),
    }
)
for column in list("abc"):
    gb = df.groupby(column)
    gb.size()
    %timeit gb.size()

# main
# 28.5 ms ± 1.34 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)
# 33.3 ms ± 1.53 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)
# 48.4 ms ± 12.5 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)

# PR
# 10.4 ms ± 498 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)
# 13.1 ms ± 63 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)
# 18.5 ms ± 44.6 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)