Fix confidence ensembles RNNT logprobs selection logic for exclude_blank scenario

[KunalDhawan]

What does this PR do ?

• Fixes the algorithm for filtering blank frames while building confidence ensembles
• Adds checks for confidence value calculation to handle the edge case of all blank frames

Collection: [Note which collection this PR will affect]
ASR

Changelog

• changes to get_filtered_logprobs() and compute_confidence() functions in nemo/collections/asr/models/confidence_ensembles.py

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

• New Feature
• Bugfix
• Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

[Kipok]

Just a small issue to fix for consistency with rnnt logic.

By the way, were you able to figure out why rnnt probabilities are not correctly normalized?

[Kipok]

Thanks!