Hi @tengfey,
Not currently. That specific visualization would need to be done outside of QIIME2.
See this thread
See this thread. For differential abundance testing, ANCOM is what you are asking for. Pairwise testing with ANCOM is one approach to compare two groups (similar to running pairwise t-tests or WIlcoxon tests).
For other metrics, e.g., alpha and beta diversity, yes. Appropriate multi-group and pairwise tests (e.g., Kruskal Wallis) are used by default — see the moving pictures tutorial for more details.
I hope that helps!