Tidy the repository: remove docs/research and bench, fix references

docs/validation/P9.md

@@ -6,7 +6,7 @@ one of which stopped after 2 min because the image lacked a C linker).
 
 ## What was built
 
-Profiling first (`bench/profile_genotype.py`): one GSA sample took 3.4 s on a Pi core, 63 % of it in calling — the
+Profiling first (a cProfile run of one sample; the benchmark scripts were retired after 2.0.0 and live in the git history): one GSA sample took 3.4 s on a Pi core, 63 % of it in calling — the
 bit-exact port evaluates `exp`/`log10`/`pow` element-wise through CPython's `math` (≈ 6.8 M scalar calls per sample) —
 and 24 % in the normalization fit's nearest-neighbour searches. Reading a manifest cost 13 s (GSA) to 53 s (Omni2.5)
 per command. The IDAT parser was already negligible (0.02 s), so it was not ported.
@@ -50,7 +50,7 @@ Manifest loading: GSA 13 s → 3 s, Omni2.5 53 s → 13 s, GDA 40 s → 10 s (Pi
 
 ### Scaling to 32 threads
 
-360 samples, `simurg genotype --no-cohort-qc`, per-sample phase from the run log (`bench/scaling.py`):
+360 samples, `simurg genotype --no-cohort-qc`, per-sample phase from the run log:
 
 | Workers | GTCs on the gp3 disk (125 MB/s) | | GTCs on a RAM disk | |
 |---|---|---|---|---|