I think Tara is on the right track by doing a deep integration with Github and Gitlab. IMHO project management tools need to track progress automatically where possible and grab relevant information related to pull request status, CI/CD feedback etc.
The Gitlab and Github issue trackers do that and aren't that bad for managing large scale projects actually. Their weakness is that they are inherently tied to a single source repository and most teams and organizations would have many of those. Fragmenting information across multiple issue trackers is not ideal. IMHO Microsoft is on to something by grabbing Github and deeply integrating it into their stack. CI/CD is where the action is and a lot of that is cloud centric and a great opportunity for upselling things like Azure. I've been keeping an eye on the github issue tracker and they've added some nice features there related to project management. There's also codespaces, project discussions (in beta currently) and a few other features that basically indicates to me that they want to be a one stop shop for anything a development team needs. An org level issue tracker that aggregates information from all source repos and their issue trackers would be a great move for them.
Jira has plugins for github integration that I've never seen working properly; most of them you need to pay for to even find out if they work or how. But the core issue is that this is clearly an afterthought for Atlassian and their tools are generally UX challenged resource hogs that are poorly integrated with each other even. They don't do github because they are still flogging the dead horse that is bitbucket; which is thankfully a lot less commonly used than gitlab and github. Basically they have a lot of me too products; I like exactly none of them. Trello used to be kind of nice and they sort of bolted it onto Jira in a weird way.
We are currently on Asana and kind of liking it but the lack of github integration is not great. What's also not great is the lack of markdown support and things like code samples. But it gets the job done and the non techies in our company like it too; which is important for us.
For a lot of projects I do, the release process these days is centered around github releases and github actions. E.g. we have internal libraries that we tag that get subsequently built and pushed to package repositories. A new release typically addresses pull requests and closes issues. It signifies the movement of those issues from doing to done (i.e. deployed/live/etc.) and reviewed (pr approved). The act of creating a branch can signal the act of starting a task. I consider work in progress pull requests a good practice for these as you get to have early feedback and discussion around it.
Any tool that forces me to do task transitions manually is fundamentally wasting my time. Releases have release documentation associated with them. Compiling a list of "what did we actually change" manually is likewise busy work you should not have to waste brain cycles on. So deep integration with tools is the next logical step for a good issue tracker. Most existing ones don't do a great job of it. That's the market opportunity.
Atlassian is ready to be disrupted. They got too big and bloated and they are not on top of their game for a long time. There are a lot of dinosaurs using them and a lot of misguided PMs insisting that the Atlassian way is THE way. IMHO they probably should but can't afford to gobble up Gitlab and that's actually a good thing. A reverse takeover would be a great lateral move for Gitlab actually. Also Google or Amazon could do worse then attempting to buy either (or both) and mirror Microsoft's strategy of owning everything relevant to running a software project in house.