In some sense, this is probably the most unsatisfactory answer possible, but it's all I know:
In the 2005 paper 1, Movahedi-Lankarani and Wells write the following in VII (p. 16):
Tomi Laakso (6) has informed us that his non-embeddability theorem [9] can be generalized in such a way that the non-embeddability theorem of Cheeger [4] no longer implies it. In a forthcoming paper, he gives a sufficient condition for a metric space to admit no bi-Lipschitz embedding into uniformly convex Banach spaces. Furthermore, he shows that there exists a metric space which does not admit a bi-Lipschitz embedding into any uniformly convex Banach space even though it admits a David-Semmes regular mapping onto $\mathbb{R}^2$.
Reference [9] is to the paper of Laakso in which he constructs Ahlfors $Q$-regular spaces of arbitrary $Q>1$ admitting Poincare inequalities and not bi-Lipschitz embedding into any uniformly convex Banach space (which was published in GAFA). Unfortunately, the footnote (6) is just ``private communication''"private communication". As far as I know, the paper"forthcoming paper" of Laakso mentioned in this quote never came out.
I for one would be happy to see an example like this, but I have not seen one anywhere else. Of course, someone might know better.
1 H. Movahedi-Lankarani and R. Wells, On bi-Lipschitz embeddings Port. Math. (N.S.) 62 (2005), no. 3, 247–268.