I don't recall any BIP which suggest to increae maximum block size by 1MB every 4 year.
Alright, then what is the plan regarding any increase of block size?
See BIP 100 - 109.
Because all I read is about side/second layers and seems there is no plan for block size change at all, as if they have abandoned that notion altogether and only focused on LN, which by the way it has come to a stale state of development without any progress.
I get your point. Roughly after 2017, there's few discussion about increasing maximum block size.
Is there any possible way for miners to identify ordinals and other types of garbage before adding them to their blocks? If there is then why won't they demand a much higher fee rate than ordinary transactions? That way not everyone will be able to clog the mempool like they are doing right now.
Yes, you just need to look for certain pattern on witness data. For example, Ordinals documentation mention string "ord" always used.
A text inscription containing the string "Hello, world!" is serialized as follows:
OP_FALSE
OP_IF
OP_PUSH "ord"
OP_PUSH 1
OP_PUSH "text/plain;charset=utf-8"
OP_PUSH 0
OP_PUSH "Hello, world!"
OP_ENDIF
First the string
ord is pushed, to disambiguate inscriptions from other uses of envelopes.