Investment casting larger, more complicated shapes is definitely possible. Catch is, if you reverse engineer from an existing piece like a Bren receiver, obtain the dimensions, and then a die sinker creates the dies to inject the waxes, which are then used to produce castings, the dimensions of the casting aren't going to be the same as the original. The product must be redesigned for production by casting. With allowances built in for machining critical areas. Making a two piece unit would likely simplify things. The castings become smaller.
In the 1980s, a simple three piece die to injection mould a synthetic stock cost $90 000 (conversation with Lee Six). Dies to produce production ready castings for a Bren reciever - and any other parts which could not be adapted from originals - would make that sum look like a pittance. Develpoment costs to produce usable Bren receivers by casting would be substantial.
One reason that limited production runs are done with CNC rather than casting is because of cost of designing and developing production ready dies.
At one time I owned 12 dies to inject the waxes for some gun parts. Investment casting is a valid technique, but it is not a gimme.
The cost to develop and produce the dies - and do the product development to get the resulting parts working - is the same for 100 or 10 000 units. Up front cost is heavy.
If a company were to take on such a project, it would be a substantial one from an industrial standpoint. It would take a serious investment of time, money and engineering talent before gun #1 appeared.
And, of course, spending dollar one before knowing if the resulting product would be classified non-restricted would be unwise. Recent tales of woe demonstrate the risks involved in production of paramilitary firearms.