RL Examples had bugs on current gym version

[sanggusti]

Your issue may already be reported! Please search on the issue tracker before creating one.

Context

• Pytorch version:
• Operating System and version: Ubuntu 20

Your Environment

• Installed using source? [yes/no]:
• Are you planning to deploy it using docker container? [yes/no]:
• Is it a CPU or GPU environment?:
• Which example are you using: reinforcement_learning
• Link to code or data to repro [if any]:

Expected Behavior

This example script (reinforce.py and actor_critic.py) should be running well without encountering any bugs.

Current Behavior

When running the script (reinforce.py and actor_critic.py), there are error:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
[<ipython-input-8-263240bbee7e>](https://localhost:8080/#) in <cell line: 1>()
----> 1 main()

[<ipython-input-4-6af08085b221>](https://localhost:8080/#) in main()
     87     running_reward = 10
     88     for i_episode in count(1):
---> 89         state, _ = env.reset()
     90         ep_reward = 0
     91         for t in range(1, 10000):  # Don't infinite loop while learning

ValueError: too many values to unpack (expected 2)

Possible Solution

Here I put my pull request that run on my system (gym version 0.25.2) https://github.com/pytorch/examples/pull/1212

Steps to Reproduce

1. Go to folder reinforcement_learning
2. run actor_critic.py or reinforce.py with gym version 0.25.2 ...

Failure Logs [if any]

### Tasks
- [ ] https://github.com/pytorch/examples/pull/1212