Summary of question:
Is it a violation of the Developer Agreement, Developer Policy, and/or Display Guidelines to choose not to display a tweet based on application logic?
The main policy section I could find is in the Developer Policy, I. Guiding Principles, 2. Maintain the Integrity of Twitter’s Products.
d. Do not modify, translate or delete a portion of the Content.
e. Maintain the features and functionality of Content and Twitter API. Do not interfere with, intercept, disrupt, filter, or disable any features of Twitter or the Twitter API, including the Content of embedded Tweets and embedded timelines.
In subsection e., you are forbidden from "filter"ing any features of Twitter, but that doesn’t necessarily mean that you cannot filter Content.
In subsection d., you are forbidden from deleting a portion of the Content, but that doesn’t necessarily mean that you cannot delete all of the Content.
The Display Guidelines has phrases such as “Display real, unmodified Tweets from real accounts.” and “Use the full text of the Tweet.”, but those don’t necessarily mean that you must display the Tweet at all.
Also relevant: TweetDeck allows you to “mute” terms or users. A “muted” term or user will cause the tweet to never display in any column.
This could be read as implicitly allowing this behavior, but TweetDeck is run by Twitter, so that’s not always a safe call to make.
The desired result is to remove Tweets from users you have blocked from Twitter Web search results.
So: Is it a violation of any relevant policies to use application logic to choose not to display a Tweet?