REF: avoid using DTA/PA methods in Week.apply_index

[jbrockmendel]

We avoid a couple of array allocations in the process.

In [2]: dti = pd.date_range("2016-01-01", periods=10000, freq="S")              
In [3]: off = pd.offsets.Week(1, False, 3)                                      
In [4]: %timeit dti + off                                                       
3.22 ms ± 30.7 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)  # <-- master
344 µs ± 4.97 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)  # <-- PR

[jreback]

sure