[RLlib] CRR documentation

[kouroshHakha]

Why are these changes needed?

Related issue number

Checks

• I've run scripts/format.sh to lint the changes in this PR.
• I've included any doc changes needed for https://docs.ray.io/en/master/.
• I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
• Testing Strategy
  • Unit tests
  • Release tests
  • This PR is not tested :(

[christy]

Nice work, Kourosh!

Need to also add link refs to crr from:

• Github readme: https://github.com/ray-project/ray/blob/master/rllib/README.rst
• RLlib docs index > Algorithms pull-down: https://docs.ray.io/en/master/rllib/index.html

New Algo Org:
Offline RL (put Off-policy Eval here): CRR, BC, MARWIL, CQL, OPE ("importance sampling" and "weighted importance sampling")
Model-free RL (split into off/on policy):
On-policy: APPO, PPO, IMPALA, A2C, A3C, PG
Off-policy: DQN, DDPG, TD3, SAC
Model-based RL: MBMPO, Dreamer
Derivative-free Algos: ES & ARS
RecSys and Bandits (or separate categories): LinUCB, LinTS, SlateQ
Multi-agent: QMIX, AlphaStar, MA-DDPG

[sven1977]

Great work @kouroshHakha , thanks for the PR.