[bug-fix] Fix issue with SAC updating too much on resume

[ervteng]

Proposed change(s)

SAC when resuming from a checkpoint would do many updates before continuing to step the environment. This is because we don't set the update_steps properly when resuming from a checkpoint, so the SAC trainer will update until it has enough update_steps.

Also fixes a bug where SAC doesn't update even after hitting buffer_init_steps.

Types of change(s)

• Bug fix
• New feature
• Code refactor
• Breaking change
• Documentation update
• Other (please describe)

Checklist

• Added tests that prove my fix is effective or that my feature works
• Updated the changelog (if applicable)
• Updated the documentation (if applicable)
• Updated the migration guide (if applicable)

[andrewcoh]