docs: fix PPO title, variable typo, baselines URL, and PQN doc URL

[mturan33]

Description

Fixes four small documentation issues:

1. Page title typo on docs/rl-algorithms/ppo.md:1: # Proximal Policy Gradient (PPO) → # Proximal Policy Optimization (PPO). The original paper (Schulman et al., 2017, arXiv:1707.06347) is titled "Proximal Policy Optimization Algorithms", and this is how PPO is universally referred to. The same page already uses the correct name on line 11. Note that PPG (Phasic Policy Gradient) is a distinct algorithm with its own page at docs/rl-algorithms/ppg.md — unrelated. The same typo also occurs in README.md, docs/rl-algorithms/overview.md, docs/rl-algorithms/ppo-rnd.md, and docs/blog/posts/cleanrl-v1.md.

2. Variable-name typo olad_approx_kl → old_approx_kl in three docs (ppo.md, ppg.md, ppo-trxl.md). Matches the variable name used throughout the code (e.g. cleanrl/ppo.py).

3. Repo-name typo openai/baselies → openai/baselines (8 occurrences in docs/rl-algorithms/ppo.md).

4. Broken URL in docs/rl-algorithms/pqn.md: link to https://iclr-blog-track.github.io/2022/03/25/pqn-implementation-details/ currently 404s. Corrected to .../ppo-implementation-details/, matching the 11+ other references in the repo to the same blog post (The 37 Implementation Details of PPO by Huang et al., 2022).

Types of changes

• Documentation

Checklist:

• I've read the CONTRIBUTION guide (required).
• I have ensured pre-commit run --all-files passes (required).
• No behavioral changes (docs-only, not performance-impacting).

[vercel]

@mturan33 is attempting to deploy a commit to the Costa Huang's projects Team on Vercel.

A member of the Team first needs to authorize it.