bench: drop sub-noise-floor sizes from bitwise_not_vortex_buffer_mut

bitwise_not_vortex_buffer_mut[128/1024/2048] was the noisiest benchmark on
CodSpeed after #8490: it flipped between the same two values in both
directions (+27.1%/-10.66% etc.) on roughly half of ~47 recently merged PRs,
including PRs changing only docs, uv.lock, or workflow YAML.

The Not impl for an owned BitBufferMut runs bitwise_unary_op_mut in place -
no allocation, no memcpy. At 128-2048 bits that is a handful of word ops, so
the reported number is fixed divan harness overhead plus binary code layout,
both of which shift with any unrelated code change. Those sizes never
measured the operation at all, so remove them outright; 16384 and 65536,
where the loop dominates, were never flagged and remain.

Signed-off-by: Claude <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01T6PPcdrqcNeUkfi1EGd4oC

Author: claude

vortex-buffer/benches/vortex_bitbuffer.rs

@@ -246,7 +246,12 @@ fn bitwise_not_vortex_buffer(bencher: Bencher, length: usize) {
         .bench_values(|buffer| !&buffer);
 }
 
-#[divan::bench(args = INPUT_SIZE)]
+/// The in-place NOT on an owned `BitBufferMut` performs no allocation and no copy, so below a
+/// few thousand bits the measurement is fixed harness overhead and binary code layout rather
+/// than the loop itself. Only the sizes where the loop dominates are worth measuring.
+const NOT_MUT_INPUT_SIZE: &[usize] = &[16_384, 65_536];
+
+#[divan::bench(args = NOT_MUT_INPUT_SIZE)]
 fn bitwise_not_vortex_buffer_mut(bencher: Bencher, length: usize) {
     bencher
         .with_inputs(|| BitBufferMut::from_iter((0..length).map(|i| i % 2 == 0)))