Drop the inventory counts from the README

It opened by telling a developer the client covers so many operations over so
many paths in so many services, and repeated variations of that further down. The
number is not something a reader can act on, it was wrong — the document had
moved on — and keeping it right means measuring the document on every build to
maintain a sentence nobody needed.

The prose says what the client IS. Where a count explained real behaviour — some
operations declare no response body, so their method hands back nothing to
decode — the explanation stays and the tally goes.

README.md

@@ -1,7 +1,7 @@
 # Hanzo Python SDK
 
 `hanzoai` is the Python client for the Hanzo API, generated from the API's own
-OpenAPI document — <!--counts-->2,259 operations over 1,620 paths, grouped into 117 services<!--/counts-->.
+OpenAPI document.
 Every `/v1` route is in it, and the names it exposes are the document's operation
 ids rather than a hand-picked subset. [`.spec-lock`](.spec-lock) names the commit
 and sha256 of the document this tree was cut from.
@@ -117,9 +117,8 @@ inside a generated client is the drift these SDKs exist to prevent. It comes bac
 the day the document describes the body.
 
 The same gap shows up in `money`, which reads its two payloads through the
-generated `*_without_preload_content` variant: of 2,259 operations, 719 declare no
-`responses` and another 117 declare no response content, so 836 of them model no
-body to deserialize. Those become ordinary typed calls when the schemas land, and
+generated `*_without_preload_content` variant: an operation that declares no
+`responses`, or a 2xx carrying no content, models no body to deserialize. Those become ordinary typed calls when the schemas land, and
 nothing else about them changes.
 
 Reference for the routes themselves: [api.hanzo.ai/docs](https://api.hanzo.ai/docs),