[BUG] `int(QuadPrecision('inf'|'nan'|'1e30'))` silently returns `INT64_MAX`

[SwayamInSync]

int(QuadPrecision('inf'))   # = 9223372036854775807   (should raise OverflowError)
int(QuadPrecision('nan'))   # = 9223372036854775807   (should raise ValueError)
int(QuadPrecision('1e30'))  # = 9223372036854775807   (should return ~1e30)

Root Cause is the following piece of code inside src/csrc/scalar_ops.cpp

return PyLong_FromLongLong(Sleef_cast_to_int64q1(self->value.sleef_value));

We should instead take the quad_to_pylong path, already present for as_integer_ratio and raise on NaN/Inf.