REF: remove is_simple_frame from json code

[jbrockmendel]

cc @WillAyd the simple_frame path is kludgy, I expect fragile to non-numpy blocks. This just rips it out to simplify the code.

[lithomas1]

Any perf impact?

[WillAyd]

lgtm if no regression. good to get rid of these branches

[jbrockmendel]

import numpy as np
import pandas as pd

arr = np.random.randn(10**4, 10)
df = pd.DataFrame(arr)

In [4]: %timeit df.to_json()
12.8 ms ± 1.72 ms per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- main
12.4 ms ± 448 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- PR

In [5]: %timeit df.T.to_json()
12 ms ± 413 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <- main
19.2 ms ± 309 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)  # <- PR

So for very-wide cases this takes a hit, but indistinguishable for normal cases.

[WillAyd]

I think that is fine. Not worth optimizing to_json for very wide frames at the ms level

[mroeschke]

Thanks @jbrockmendel