Throw a clear error for degenerate-length data vectors (fixes #121)

[JohnCobbler]

Give quantile! a clear ArgumentError instead of an internal AssertionError
for a data vector that reports length 0 while not being empty, fixing #121.

An integer range whose length overflows (e.g. typemin(Int):typemax(Int)) has
length == 0 but isempty == false. Such input passes the isempty guard in
_quantilesort! and then trips @assert n > 0 inside _quantile, whose comment
claims the case "should never happen here".

This adds a length(v) == 0 guard in _quantilesort! (covering both quantile!
entry points before any _quantile call) that throws an ArgumentError pointing
the user at the overflow and suggesting they collect to a concrete vector first.
The now-redundant @assert in _quantile becomes a defensive ArgumentError,
which also keeps the check under --check-bounds=no where @assert may be elided.

r = typemin(Int):typemax(Int)
quantile!([0], r, [1])        # ArgumentError, was AssertionError
quantile(r, 0.5; sorted=true) # ArgumentError, was BoundsError

The default quantile(r, 0.5) path is unchanged: it materializes the range via
Base.copymutable first, which errors before reaching this code; that is a
separate, appropriate failure for a range claiming 2^64 elements.

Tests cover both overflow paths and confirm the existing empty-vector behavior
(ArgumentError("empty data vector")) still fires.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.44%. Comparing base (e14c8f8) to head (06b7ad1).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #206   +/-   ##
=======================================
  Coverage   96.43%   96.44%           
=======================================
  Files           2        2           
  Lines         449      450    +1     
=======================================
+ Hits          433      434    +1     
  Misses         16       16           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.