Harden sprint4 CLI solvers against input-driven DoS and unbounded allocation

[krotname]

Motivation

• Prevent denial-of-service from attacker-controlled stdin sizes and predictable hash collisions in the sprint4 CLIs by adding bounds and unpredictability.
• Avoid unbounded memory or CPU growth when these utilities are invoked on untrusted input.

Description

• For Map: add a per-process random HASH_SEED and mix keys before bucket selection to mitigate attacker-controlled collision chains, add MAX_COMMANDS, replace accumulated StringBuilder output with streamed BufferedWriter, add bounded and validated integer parsing (nextInt(min,max)), validate commands, and move CLI logic into solve() wrapped by a safe main that rejects invalid/excessive input.
• For FindSystem: introduce MAX_DOCUMENTS, MAX_QUERIES, and MAX_LINE_LENGTH, enforce these limits in nextInt(max) and nextLine(maxLength), stream per-query output with BufferedWriter, and catch IOException in main so oversized or malformed input is rejected without exhausting resources.
• Keep original algorithms and data structures (chained hash table and inverted-index logic) but add defensive input handling and a randomized hash mapping to preserve functionality while reducing attack surface.

Testing

• Compiled both classes with javac -d /tmp/cwjclasses src/main/java/algorithms/sprint4/Map.java src/main/java/algorithms/sprint4/FindSystem.java and compilation succeeded.
• Exercised CLIs with representative inputs for algorithms.sprint4.Map and algorithms.sprint4.FindSystem and observed expected outputs.
• Verified oversized/malformed inputs are rejected (e.g. very large n now triggers a controlled IOException rather than exhausting heap).
• Ran the full test suite with mvn test which completed successfully (546 tests run, 0 failures/errors).

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d47ca83d2e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[krotname]

Закрываю как superseded by #116: изменения перенесены в одну проверенную ветку, все review findings учтены; локальный mvn verify прошёл полностью.