Tokio 1.44+ (tokio-rs/tokio#7880) shortened the default worker thread
  name from  to . The default
  prefix discovery now tries both the old and new names so step (b) can
  short-circuit without falling through to the 500ms stack sampling.

Add 9 unit tests for `is_blocking_pool_stack()` covering the blocking
pool detection logic from issue #3. This function went through multiple
release iterations (v0.4.2–v0.5.0) but had zero test coverage.

Fix TROUBLESHOOTING.md listing 3 worker discovery steps when the actual
implementation and ARCHITECTURE.md document 4. Fix README.md stating
"x86_64 architecture" when ARCHITECTURE.md and DEVELOPMENT.md both
document x86_64/aarch64 support.

// ticktockbent

get_cpu_id() was stubbed to always return 0 because aya-ebpf didn't
appear to expose bpf_get_smp_processor_id. Turns out it does, via the
re-exported bindings (helper #8), it's just #[doc(hidden)] so it
doesn't show up in the generated docs.

Every exported trace event was reporting cpu_id: 0 regardless of which
core the sample was actually taken on. The TUI doesn't render per-CPU
data so nobody noticed, but the Chrome Trace JSON was silently wrong.

// ticktockbent

Add unit tests for blocking pool filter, fix doc inconsistencies

Fix get_cpu_id() to return actual CPU core instead of hardcoded 0