Perhaps, this is a meta-level question. Most review papers narrate the story of the field/problem or summarize a large but focused body of literature. The authors are usually senior level faculties with years of experience in the field. How these papers are being reviewed and what is the formalism behind accepting or rejecting them? Many such papers fail to present a framework/criteria/dimension for the field (maybe due to complexity), but is there any effort targeted to develop reviewing skills for young researchers to follow?
It is true that most publications leave a new graduate student without a sufficient introduction to the topic; however, this should not be the case with a Review. As a grad student, I recall being frustrated reading regular journal articles. Then I remember my elation while reading the introduction of my first review article. The clouds of confusion parted as the author explained few-cycle laser pulses in "plain physics."
Review articles are arguable the hardest to write. First, the author must be intimately familiar with the topic and be able to weave a disparate group publications into a cohesive story. The Review's introduction must present sufficient motivation, provide a historical context, and create a structural outline for the topic at hand. Next, the author must clarify and distill the important papers, and include lesser papers as they advance the story. Since Reviews often include over 200 articles in their reference list, this can be quite a task. Finally, a good Review will highlight new applications and open question.
In addition to the above structural criteria by which a Review should be judged, see the selected answer of this post for political advice.
To summarize and answer your question:
Why don't junior scientists write such Reviews, and why aren't there efforts train them?
Before learning to run one must learn to walk, and before writing a Review one must learn to write a regular article. As you mentioned, senior faculty often write Reviews because they have the most familiarity with the literature, connections, and experience. The best training for junior faculty is gaining a familiarity with: the literature, the publishing process, and colleagues.
I too find it frustrating that so many literature reviews are so unsystematic.
There are efforts underway to improve the quality of reviews, and to develop these skills in early-career researchers. We like our students, where possible, to write a review article as a early part of their PhD. The medical field has for many years been honing the quality of reviews, and there's much that other fields can learn from it.
We have had a Q&A about systematic reviews earlier, and there I listed three relevant books. The most significant of these, as far as your question is concerned, is Doing a Systematic Review: A Student's Guide - Angela Boland et al - it's primarily aimed at Masters students, and also makes for an easy-to-read introduction to the field for anyone.