I did not want to focus on the example as it was but one example of common user behavior, but let's do it anyway. I should preface this by saying that I have been a moderator on online spaces for the last two decades. First starting on BBS, IRC, and more recently social networks. Social networks contain some of the worst behaviors I have ever experienced.
It is not words that need to be banned, banning words would be useless. Users could simply call people "gr00mers". Another way people use to talk about trans people in a hateful way is "troons", which is defined on Urban Dictionary as:
Troon, Noun. A transphobic slur referring to a trans woman.
The word Troon used to mean "an ugly woman who looks like a man." However, over time the word has been co-opted and turned into a particularly nasty slur against trans women.
This word was quickly identified in groups and people would get removed from groups for using it in an abusive way by setting that word to trigger an moderation warning. The people using it quickly adapted, proving that banning words is useless. This changed to "trains", which was added to the watch list. The little dance kept going on, right now the way it is written is by using a train emoji. So you would see "Look at this filthy[train emoji]".
Banning words is useless. Words are not hardcoded constants that cannot be replaced, they are descriptive and evolve in meaning. Banning "groomer" would simply lead to people using another word. Most slurs are just that, words that are borrowed and then infused with hate. Just take a look at https://en.wikipedia.org/wiki/List_of_ethnic_slurs, you will find that a lot of the terms are harmless when they are not weaponized or used in a hateful context. Even one of the oldest word used to indicated an intolerant position toward transgender people was simply a way to write "[a car's] transmission", or "bundle of sticks" towards homosexual people. This list is not exhaustive at all and only community websites such as Urban Dictionary is able to track slurs as the list grows constantly.
What I am talking about is seeing through the thing veil of false politeness and banning harmful actions.
Adding to the previous example: people often come ("raid") in private Facebook support groups. To enter those groups, you need to answer several questions. For example, "Are you transgender?/Do you understand that this is a support group and that confrontation is frowned upon?/Do you agree with the rules of the group?". Every single day, people lie to those questions to enter. Upon entering, they quickly drop all false pretense to quickly find the most vulnerable of users and spam them with hate via their inbox (where moderators cannot keep a watch). On threads about depression, they leave comments encouraging the person to end their lives through words that are absolutely legal. Imagine the worst 4chan has to offer but directly in your inbox. Many users are not web-savvy and this type of content will be absolutely crushing. Worst of all, trolls gather in groups where the only purpose is to list the profile of the vulnerable users they find. Some of those groups that I am aware of have up to 25k members. This is worrisome to me. Since they only contain links to profiles and are worded using legal ways, Facebook does not shut them down. They result in users receiving hate and thinly veiled threats from up to a hundred users a day. Since those message are worded in a way as to avoid reports from sticking.
Trans issues are a hot and controversial topic, so let's take a step back and use another community as an example.
Imagine you ran a community of people with anorexia gathering to talk about treatments, solutions and review doctors. One facet of anorexia is body dysmorphia where the person sees themselves as very overweight even when dangerously underweight. The hateful person or troll can find someone on door's death and convince them that they are fat. Is it ethical? Not at all. But if worded by someone skilled at trolling, it can be done in a legal way. They can send that person image of fat people and tell them that they look the same. Another way they go through that is by undermining the authority of doctors and hinting towards false and often dangerous treatments. This kind of trolling will go on for weeks, even months. With people advocating strange concepts under a false account with a false profile photo. Some of them even staying undercover for so long that they manage to gain a high standing in those community only to then use the insider information they have gathered to nefarious purposes.
Often, the trolls do not even hold a grudge or hate towards the targeted communities. To them, it is a game. Many are very young teens acting under the impression that they are great hackers for managing to infiltrate support networks and destroying them.
We can all agree that those behaviors are not only hateful, they are dangerous and target the most vulnerable of individuals. Worst of all for the corporations running those websites, they result in huge exodus of users. One I can think of is a 40k strong Facebook group that moved to Discord due to the constant abuse their members received.
Can Twitter decide to accept the behavior described above? Absolutely. It is a private entity. However this kind of "everything is allowed" at all cost mindset comes with a cost. Often, this cost is the migration of users to other platforms. What happened to Tumblr is a great example.