Special Relativity and Gravity As Einstein was seeking a relativistic theory of gravity, he thought that special relativity should be upgraded to general relativity thus promoting the Minkowski space to curved pseudo-Riemannian (Lorentzian) one. Does this mean that special relativity as a theory never discussed gravity from any perspective?
 A: 
Does this mean that special relativity as a theory never discussed gravity from any perspective?

It all hinges on the luminiferous aether which was prevalent in the 19th century theories:
The Michelson Morley experiment was crucial in discovering that there does not exist a luminiferous aether.

The Michelson–Morley experiment was performed over the spring and summer of 1887 by Albert A. Michelson and Edward W. Morley at what is now Case Western Reserve University in Cleveland, Ohio, and published in November of the same year. It compared the speed of light in perpendicular directions, in an attempt to detect the relative motion of matter through the stationary luminiferous aether ("aether wind"). The negative results are generally considered to be the first strong evidence against the then-prevalent aether theory, and initiated a line of research that eventually led to special relativity

To start with the Lorenz  transformations were discovered/invented to make consistent Maxwells equations with the existence of a luminiferous ether, i.e. an inertial framework against which everything else would be moving with classical mechanics equations of motion.
Here is the history of Lorenz transformations, the lynch pin of special relativity.

Lorentz (1892–1904) and Larmor (1897–1900), who believed the luminiferous ether hypothesis, were also seeking the transformation under which Maxwell's equations are invariant when transformed from the ether to a moving frame. They extended the FitzGerald–Lorentz contraction hypothesis and found out that the time coordinate has to be modified as well ("local time"). Henri Poincaré gave a physical interpretation to local time (to first order in v/c) as the consequence of clock synchronization, under the assumption that the speed of light is constant in moving frames. Larmor is credited to have been the first to understand the crucial time dilation property inherent in his equations.
In 1905, Poincaré was the first to recognize that the transformation has the properties of a mathematical group, and named it after Lorentz.
Later in the same year Albert Einstein published what is now called special relativity, by deriving the Lorentz transformation under the assumptions of the principle of relativity and the constancy of the speed of light in any inertial reference frame, and by abandoning the mechanical aether.

The "out of the box" thinking of Einstein comes when he applied the Lorenz transformations to particles, not light. It took some time to confirm it , and the real validation comes from nuclear physics and the huge number of particle physics experiments which can only be interpreted by assuming a four dimensional space time.
As you see from the above precis gravity does not enter into the special relativity validation.,
A: General relativity was not the first gravitational theory proposed for special relativity. The first steps were just to treat it like any field theory (like it is in classical physics). 
The most basic is the scalar field theory, which has been proposed in some variations by Einstein, Abraham and Nordström : 
\begin{equation}
\square \Phi = 4 \pi G \rho
\end{equation}
Which is just the Newtonian Poisson equation with $\Delta \rightarrow \square$. To remain Lorentz invariant, $\rho$ is not the energy density but the rest mass density.
Nordström also made later on the following theory : 
\begin{equation}
\Phi \square \Phi = -4 \pi G T
\end{equation}
With $T$ the trace of the stress energy tensor. 
While both agree with the Newton equation in the classical limit, they has a few problems. Light is not deflected by gravity (since it has no rest mass or a trace of the stress energy tensor), and the precession for Mercury is of the wrong sign and magnitude. It is also a bit odd to put into Lagrangian form, since the stress energy tensor is directly in it.
Another proposed theory was the tensor theory, which was 
\begin{equation}
\frac 1 2 \square h_{\mu\nu} + \partial_\mu \partial_\nu h - \partial_{\{\nu}\partial^\sigma h_{\mu\}\sigma} + \eta_{\mu\nu} (\partial^\alpha\partial^\beta h_{\alpha\beta} - \square h) = \frac \kappa 2 T_{\mu\nu}
\end{equation}
Which is basically the same as linearized GR. Due to this, it has a pretty good agreement for experimental values, but it has theoretical problems that if the matter fields are on shell, energy is not conserved any more. To fix this, the procedure is to add more and more counter terms, which will leave you with general relativity in the end.
A: Einstein considered a theory of gravity involving a scalar field in a flat spacetime. Einstein discarded this theory because he thought it broke conservation of energy. Domenico Giulini has claimed that Einstein's reason for discarding the theory was wrong but the theory conflicts with experiment anyway:
https://arxiv.org/abs/gr-qc/0611100
A: see H.Poincaré  1905 Sur la dynamique de l’électron
his § 9. — Hypotheses on gravitation is very, very, on top of what you are asking.  
IMO, his document is much more pertinent to Relativity than the Einstein's one and he deals with gravity under SR principles. He could not make it complete because he did not know the speed of the solar system thru the background, as he confessed. (He needed one more equation to close the Mercury' issue).  
