ENH: Preserve input start/end type in interval_range

[mroeschke]

In [1]: import pandas as pd

In [2]: import numpy as np

In [3]: pd.interval_range(start=np.int8(1), end=np.int8(10))
Out[3]: 
IntervalIndex([(1, 2], (2, 3], (3, 4], (4, 5], (5, 6], (6, 7], (7, 8], (8, 9],
               (9, 10]],
              dtype='interval[int64, right]')

It would be ideal if the resulting dtype was 'interval[int8, right]'

[VISWESWARAN1998]

In the interval.py, I made it retain the same datatype if both start and end has same datatype

dtype = start.dtype if start.dtype == end.dtype else np.dtype("int64")
breaks = maybe_downcast_numeric(
    breaks,  # type: ignore[arg-type]
    dtype,
)
return IntervalIndex.from_breaks(breaks, name=name, closed=closed, dtype=IntervalDtype(subtype=dtype, closed=closed))

And I am getting the desired results