Reverse Engineering (RE) Learning Path.

Of course there are different asm(s) for different hardware platforms (ie x86, ARM, 68K etc), these are the different chips and have very big differences in the number and type of commands they support.

For pc's you only really need to consider x86 (or IA64 AMD64, an extension of x86 if you want to go for 64bit).

I guess the different "flavours" you mean are the different assemblers available for translating assembler into machine code.

For the PC (Windows or Linux or any other machine using the Intel x86 processor) there are two (3?) main ways of writing assembler, AT&T, Intel (and HLA).

The first two are the main ones, with the Intel syntax the one you will see most often.
As Intel originally designed the chips I guess it is only right they dictate the mnemonics used. Most assemblers on both windows and Linux will be able to understand this way of writing code. (ie MASM, TASM)

The Intel syntax is primarily used with the assembler which GCC uses, called GAS on the Linux platform.

HLA (High Level Assembler) is another way of writing code, although it has more features of a high-level language like C or Pascal, and can be a good place to start.

As to your op, I think that a basic grounding in the concepts of proggramming languages will always stand you in good stead when delving into the inner workings of processors, and has the benefit of more tangible results when you are starting off. Then you can transfer the concepts of variables/flow-control etc, to assembler, as it can be a bit dry and hard going when you start off.

B

PS
wikipedia is a great place to find a lot of background on this stuff, before you dive in =)

http://en.wikipedia.org/wiki/X86_assembly_language

and dont forget to get the info straight from the horses mouth so to speak, with the full set of Intel manuals

http://www.intel.com/products/processor/manuals/