I have been exploring these nowadays. I will give two reasons for why Bb5 is recommended for white and is beginning to overtake d4 as the main line.
First, unlike 2...d6 Sicilians, with 2...Nc6 Sicilians, Black can dictate variations if White enters an open Sicilian. To see this, we must start at Black's second move. After e4 c5; Nf3 d6; d4 cxd4; Nxd4 Nf6; Nf3, Black's a6 (the Najdorff) is not forcing anything on White who now has a million decent options to choose from: h3, g4, f3, a4, Rg1, Bg5, Be3, Bc4, etc. With Black playing 2...Nc6 though, there are far fewer options. Because this is a more active and developing move, unlike 2...d6, after random moves by White balck easily equalizes. After e4 c5; Nf3 Nc6, if White enters an open Sicilian (i.e. d4) then Black can force a Sveshnikov or White deviates early on from theory, which is a concession then and Black is fine in most sidelines. So, white should be prepared to enter a long sequence of theory lines as soon as he/she puts a pawn on d4 on move three. In mainline Sveshnikov, although Black's position looks scary it is proven time and again to be playable with a good fighting chance for a win.
As an e4 player you already spend half your life on all types of responses to e4 (take Najdorff itself for example). So, you begin to wonder why to enter such forced line against an opponent who is probably better equipped with theory and experience.
Once you decide to ditch a third move d4, then there are basically two objectively good moves for white: Bb5 and c3, with former being much stronger.
So, my second reason is Bb5 is indeed positionally a very logical and strong move. It is easy to see why. 1) Black's Nc6 is a great piece putting pressure on center, specially e5 and d4. So, it is good to make it uncomfortable. 2) Bishop develops to an active square and we are ready to castle, while Black has not touched their kingside, so we will be ahead in development. 3) The threat to double Black's pawns is serious and often Qc7 or Qb6 are too slow to avoid this.