Is there a standard methodology to create "cross platform FPGA desigs"?

When we create FPGA designs, we often use vendor IP. The IP requires that vendor's libraries to simulate and that vendor's compiler and device to be useable. It will not be identified with libraries, compiler and device from another vendor.

Lets take an e.g Altera has DDR3 controller IP, floating point multiplier IP, FIFOs e.t.c, when we use them, the program generates files containing all the necessary information to help us simualte and synthesize these things. However, what do we do if we want our design to work across another vendor e.g Xilinx or Microsemi?

Vendor's would want that we use only their FPGAs so they will probably make it harder to migrate to another vendor. However, I am sure that they will provide method to migrate from another vendor to their own devices to the process easier and thus get more customers to shift around.

Are there any standard tools and methodologies that can help us with this?

Unless your design is fairly simple, you are never going to create a totally vendor independent design. At some point, you're going to have something that is proprietary to the chip. All you can do is attempt to minimise this.
1. Only use a single vendor
2. Write your code to infer the basics (like bram, DSPs). Rams and DSPs are fairly common components, and the code to infer them is basically the same across vendors.
3. Create a common wrapper, with vendor specific libraries underneath. This way the code at the top can be the same, it just takes the blocks from the included libraries.
4. Just use vendor IP. You can't always escape it.
5. Write your own ip that uses the above methodologies (but expect to spend a lot of time maintaining it)
6. Documentation documentation documentation. Make sure everyone knows why (not how) you did the things that you did.