[DOCS] Troubleshooting Steps for MPI Operator

[10sharmashivam]

References Issue- flyteorg/flyte#4464
Reference PR- flyteorg/flyte#5857

This PR addresses the troubleshooting steps for the MPI operator as suggested in the feedback in [Reference PR]. I shifted the MPI common issues and troubleshooting steps for them to MPI Guide itself.

This is a documentation update based on [Issue in flyteorg/flyte ] and [Flyte Slack's conversation] about MPI operator installation issues.

The suggestions to move the troubleshooting content came from the review of my previous PR in the flyteorg/flyte repository, which is linked here: [PR MPI Operator].

Why are the changes needed?
These changes provide important troubleshooting steps for users setting up the MPI operator for distributed training jobs on Flyte. The steps cover common issues like insufficient resource allocation for worker pods and clarify the correct workflow registration method, especially when using custom images. This will help users avoid common pitfalls and streamline the MPI operator setup process.

What changes were proposed in this pull request?
This pull request adds troubleshooting steps to the MPI operator installation documentation. Specifically, the following issues are addressed:

• Insufficient Resources for Worker Pods: Steps to ensure sufficient CPU and memory requests are set.
• Workflow Registration Method Errors: Clarification on using the correct registration method when working with custom images, with a link to the relevant Flyte documentation.

Check all the applicable boxes
I updated the documentation accordingly.
All new and existing tests passed.
All commits are signed-off.

[10sharmashivam]

@davidmirror-ops, I’ve submitted a new PR addressing the suggestions made in the previous review for PR. Waiting for your thoughts on it!

[10sharmashivam]

Docs Link

https://flyte--1756.org.readthedocs.build/projects/cookbook/en/1756/auto_examples/kfmpi_plugin/index.html

[davidmirror-ops]

Thank you!