Performance/bounding box query

[MeyerBender]

Avoided unnecessary calls of compute(). On the Xenium example dataset, this yields the following improvement:

Xenium datasets (12.1M points), checked with %%timeit

• PyPi version: 4.31 s ± 85.8 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
• New version: 2.54 s ± 77.5 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)