chore: keep the 48 MB ATP dataset out of the published crate

`cargo publish --dry-run` packaged 48.1 MiB (8.6 MiB compressed) for a library
whose source is 312 KB. All of it was examples/atp.csv, a tennis dataset the
atp example reads.

examples/atp.rs opens it by relative path at runtime rather than include_str!,
so excluding the data still compiles and `cargo package --verify` still builds
every target — the example just needs the file fetched from the repo to run.

Packaged size is now 360.8 KiB / 83.1 KiB compressed, a 133x reduction. Every
consumer would otherwise have paid that download.

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

Cargo.toml

@@ -17,7 +17,17 @@ categories = ["algorithms", "science", "game-development"]
 # 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.
-exclude = ["/docs", "/benches/*.txt", "/temp", "/.gitea"]
+# `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
+# example just needs the file fetched from the repo to run.
+exclude = [
+    "/docs",
+    "/benches/*.txt",
+    "/temp",
+    "/.gitea",
+    "/examples/atp.csv",
+]
 
 [lib]
 bench = false