docs(cargo): correct the licence note — kellnr does not require one

The note claimed `cargo publish` rejects a crate without `license`. That is
true only for crates.io; publishing to an alternative registry does not check
it, verified by a dry run against kellnr that packages and verifies cleanly.

Staying unlicensed is a deliberate choice, so the note now says that and states
the actual consequence — all-rights-reserved by default — rather than a
mechanical blocker that does not exist.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01T5SYDExxL4vZgvunrcNSMc

Cargo.toml

@@ -13,10 +13,17 @@ publish = ["kellnr"]
 readme = "README.md"
 keywords = ["trueskill", "rating", "bayesian", "elo", "skill"]
 categories = ["algorithms", "science", "game-development"]
-# TODO: pick a licence before publishing. `cargo publish` rejects a crate
-# without `license` (or `license-file`), and without one the source carries no
-# stated terms. The Rust convention is `license = "MIT OR Apache-2.0"` plus the
-# matching LICENSE-MIT / LICENSE-APACHE files.
+# Deliberately unlicensed for now — not an oversight.
+#
+# Correcting what this note used to claim: `cargo publish` does NOT reject a
+# crate without `license`/`license-file` when the target is an alternative
+# registry. Only crates.io enforces it. Verified by `cargo publish --dry-run`
+# against kellnr, which packages and verifies cleanly as-is.
+#
+# The consequence is legal, not mechanical: with no stated terms the source is
+# all-rights-reserved by default. The Rust convention when one is wanted is
+# `license = "MIT OR Apache-2.0"` plus matching LICENSE-MIT / LICENSE-APACHE
+# files.
 # `examples/atp.csv` is a 48 MB tennis dataset — 99% of the packaged crate,
 # for a library whose source is 312 KB. `examples/atp.rs` opens it by
 # relative path at runtime, so excluding the data still compiles; the